This page is part of the FHIR Specification (v5.0.0-draft-final: Final QA Preview for R5 - 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

6.1.1 Security Labels

Security Work Group

Maturity Level: 3

Standards Status: Trial Use

A security label is a concept attached to a resource or bundle that provides specific security metadata about the information it is fixed to. The Access Control decision engine uses the security label together with any provenance resources associated with the resource and other metadata (e.g. the resource type, resource contents, etc.) to

• approve read, change, and other operations
• determine what resources can be returned
• determine what handling caveats must be conveyed with the data

Security Labels enable more data to flow as they enable policy fragments to accompany the resource data.

The intent of a security label is that the recipient of resources or bundles with security-tags is obligated to enforce the handling caveats of the tags and carry the security labels forward as appropriate.

Security labels are only a device to connect specific resources, bundles, or operations to a wider security framework; a full set of policy and consent statements and their consequent obligations is needed to give the labels meaning. Because of this, security labels are most effective in fully trusted environments - that is, where all trading partners have agreed to abide by them in a Mutual Trust Framework. Note also that security labels support policy, and specific tagging of individual resources is not always required to implement policy correctly.

In the absence of this kind of pre-agreement, Security Labels may still be used by individual parties to assist with security role checking, but they might not all be recognized and enforced, which in turn limits what information can flow.

Local agreements and implementation profiles for the use security labels should describe how the security labels connect to the relevant consent and policy statements, and in particular:

• Which Security Labels are able to be used
• What to do if a resource has an unrecognized security label on it
• Authoring obligations around security labels
• Operational implications of security labels

This specification defines a basic set of labels for the most common use cases trading partners, and a wider array of security labels that allow much finer grained management of the information.

6.1.1.1 Representing Security Labels

A security label is represented as a Coding, with the following important properties:

<Patient xmlns="http://hl7.org/fhir">
  <meta>
    <security>
      <system value="http://terminology.hl7.org/CodeSystem/v3-Confidentiality"/>
      <code value="R"/>
      <display value="Restricted"/>
    </security>
  </meta>
...  [snip] ...
</Patient>

{
  "resourceType" : "Bundle",
  "type" : "searchset",
  ... other headers etc.....
  "entry" : [
     ... other entries ....
     {
       "resource": {
         "id" : "1",
         "meta" : {
           "security" : [{
             "system" : "http://terminology.hl7.org/CodeSystem/v3-ActCode",
             "code" : "DELAU",
             "display" : "delete after use"
           }]
         }
         ... other content etc.....
       }
     },
     ... other entries ....
  ]
}

Note: the actual terms used in these examples are described below.

The basic framework for security labels is described by the HL7 Healthcare Classification System . This specification identifies how security labels are defined and provides a relatively comprehensive list of labels. All the HCS defined labels (see below for the lists) can be used as security labels on FHIR resources and bundles (e.g. requests and responses).

In addition, other security labels not defined here or in the HCS can be defined by jurisdictions, vendors and/or projects and used as appropriate. However, note that:

• Defining additional security labels will increase costs associated with information and system portability
• Implementation guides and applications SHOULD always use the applicable label defined by the HCS if one exists

Note: The use of security labels and the expression of common shared security policies is a matter of ongoing discussion and development in several communities.

6.1.1.2 Core Security Labels

This specification defines a set of core security labels for all FHIR systems. All conformant FHIR Applications SHOULD use these labels where appropriate. For all these labels, how they are operationalized - their use and interpretation - is subject to the applicable Mutual Trust Framework agreement as described above.

The Security Label vocabulary has three patterns of use: (1) Bundle: Security/Privacy considerations of a data set, (2) Context: Describe security/privacy context of a request for data, and (3) Meta Data: to indicate security/privacy meta about that data.

Bundle: A bundle is a container for a collection of data. Some uses of bundle are for communicating search results, sending data, or persisting data (See Bundle). The Bundle would carry meta about the data contained in the bundle. Specifically the confidentiality 'high water" mark, the authorized purposeOfUse, the required compartment clearance, Refrain, and Obligations that must be maintained. Where the "high water" mark is an indication of the most high confidentiality. Depending on Policy, the Bundle might include the cross-section of sensitivity or integrity, although this is usually not included. Provided in the Security Label Event Examples valueSet contains of some security-label codes that would be used on Bundle.meta.security.

Context: Requests (e.g. Read, Query, message triggers) - would describe the context of the request using purposeOfUse and compartment/clearance. The request might declare the highest confidentiality desired. It is unlikely to see in a request a declaration of sensitivity or integrity. It is also unlikely to see Obligations within a Request. (See Bundle for Response, where these are appropriate)

Meta Data: All resources have a meta.security element to hold descriptions (meta) about the data relative to privacy and security risk. Thus data may be tagged with confidentiality, sensitivity, and integrity. The data might be tagged with the indication of collection context using compartment or purposeOfUse. Data would not typically be tagged with Refrain, or Obligations. Provided in the Security Label Data Examples valueSet contains of some security-label codes that would be used on data.

More complex use of tagging in the data resource, bundle, context, or in Provenance, is possible.

6.1.1.3 Break the Glass

There is a special security label to support the commonly encountered "break-the-glass" protocol, where a clinician (usually in an emergency context) requests emergency unauthorized access to the patient's record.

This purpose of use label is represented as a security label on the request, rather than on a resource, and so is represented in the request as a web category :

HTTP/1.1 GET fhir/Patient/482735/condition
Content-Type: text/xml
Access-Control-Allow-Origin: *
Last-Modified: Thu, 19 Nov 2013 07:07:32 +1100
ETag: 24
Category: http://terminology.hl7.org/CodeSystem/v3-ActReason#BTG; scheme="http://hl7.org/fhir/tag/security"; label="break the glass"

While the principle of break-the-glass is well understood, implementing it well has some challenges. This specification defines a method to represent break-the-glass in an HTTP request, but does not define any policy or protocol around such requests. At a minimum, implementations must ensure:

• How, when and why to initiate the break-the-glass is well understood
• Appropriate authorization, consent checking, and access control is used to ensure it is used properly (e.g. if using OAuth, checking that the Authorization Server allows this)
• Any use is well-represented in an AuditEvent break the glass example.
• Note an example OperationOutcome that might be used to indicate potential for break-glass.

See this paper for discussion of the issues involved in break-the-glass operations.

6.1.1.4 Healthcare Privacy and Security Classification System (HCS)

The security labels described above are a subset of the full set of security labels defined by the HL7 Healthcare Privacy and Security Classification System . Note the use of "security label" is used broadly in FHIR to for all security tags. There is a more formal definition in the security labeling community of "security label" that refers to an overall assessment, in HL7 this would be represented with a value from the ConfidentialityCode value set. For more detailed on how to implement security tagging and labeling, HL7 has published Data Segmentation for Privacy Implementation Guide for FHIR

Type of security metadata observation made about the classification of an IT resource (data, information object, service, or system capability), which may be used to make access control decisions. Security classification is defined by ISO/IEC 2382-8:1998(E/F)/ T-REC-X.812-1995 as: "the determination of which specific degree of protection against access the data or information requires, together with a designation of that degree of protection."

Security classification metadata is based on an analysis of applicable policies and the risk of financial, reputational, or other harm that could result from unauthorized disclosure.

Tag Set	Car.	Description	Example Tags
Confidentiality	1..1	Security label metadata classifying an IT resource (clinical fact, data, information object, service, or system capability) according to its level of sensitivity, which is based on an analysis of applicable privacy policies and the risk of financial, reputational, or other harm to an individual or entity that could result if made available or disclosed to unauthorized individuals, entities, or processes.	Unrestricted, Normal, Very Restricted

6.1.1.4.0.1 Security Category

Type of security metadata observation made about the category of an IT resource (data, information object, service, or system capability), which may be used to make access control decisions. Security category metadata is defined by ISO/IEC 2382-8:1998(E/F)/ T-REC-X.812-1995 as: "a nonhierarchical grouping of sensitive information used to control access to data more finely than with hierarchical security classification alone."

Tag Set	Car.	Description	Example Tags
Policy	0..1	Security label metadata that segments an IT resource by conveying a mandate, obligation, requirement, rule, or expectation relating to its privacy.
Sensitivity	0..*	Security label metadata that segments an IT resource by categorizing the value, importance, and vulnerability of an IT resource perceived as undesirable to share.	STD, HIV, SUD
Compartment	0..*	Security label metadata that segments an IT resource by indicating that access and use is restricted to members of a defined community or project.	Care Team, Research Project
Integrity	0..*	Security label metadata that segments an IT resource by conveying the completeness, veracity, reliability, trustworthiness, and provenance of an IT resource.	Anonymized, Digitally signed
Provenance	0..*	Security label metadata that segments an IT resource by conveying the provenance of the IT resource's asserted or reported source.	Patient reported, Clinician asserted
Trust	0..*	Security label metadata that segments an IT resource by conveying the basis for trusting the source.	Trust Accreditation, Trust Agreement

6.1.1.4.0.2 Security Control

Type of security metadata observation made about the control of an IT resource (data, information object, service, or system capability), which may be used to make access control decisions. Security control metadata conveys instructions for secure distribution, transmission, storage or use.

Tag Set	Car.	Description	Example Tags
Purpose of Use	0..*	Security label metadata that segments an IT resource by conveying the reason for performing one or more operations on information, which may be permitted by source system's security policy in accordance with one or more privacy policies and consent directives.	Treatment, Payment, Operation, Research
General Purpose of Use	0..*	Security label metadata that segments an IT resource by conveying the reason for performing one or more operations on information of purpose of use at a general level.	Coverage, Patient Requested, Emergency Treatment
Obligation **	0..*	Security label metadata that segments an IT resource by conveying the mandated workflow action that an information custodian, receiver, or user must perform.	Encrypt, mask, comply with policy
Refrain **	0..*	Security label metadata that segments an IT resource by conveying actions which an information custodian, receiver, or user is not permitted to perform unless otherwise authorized or permitted under specified circumstances.	Do not disclose without consent, no reuse

(**) Privacy-revealing Obligation or Refrain tags such as the Obligation Policy MASK (mask) or the Refrain Policy NODSCLCD (no disclosure without consent directive), shall not be included in the high-watermark labels of a Bundle, DocumentReference, or Message Resources.

Other Value-sets are defined by the FHIR DS4P IG .