Submitted By: Henry Wei / Google | |
---|---|
Data Element Information | |
Rationale for Separate Consideration | FHIR Identifier Data Types already exist; this is not a separate element, but rather a need for better specification of existing elements that may be too ambiguous for critical interoperability and portability use cases. Within most elements e.g. Patient, there are components that often tie to organizations. This includes: *.identifier.assigner (where * is individuals such as a patient, a care team member, or relatedPerson) Provenance (often linked to an organization that is cannot only narrowly be defined by NPIs, as there are thousands of non-NPI organizations such as ACO networks, payors, community groups, and public health entities). The logical outgrowth of both portability and value-based care policy is thus a need to ensure cross-organization identifiers. One pragmatic way of doing this is to ensure the Organizational identifiers are unique, well-enumerated and properly described in the Components of the Elements, in the same way that friends in casual conversation could use the same identifier, but give context to provide unique identities across organizations, e.g. "Larry Bird from the Celtics" vs "Larry Bird from Nephrology" (one is a legendary basketball player, the other is an actual nephrologist). In this example, clear, unique Organizational identity helps individual matching by avoiding mis-matching. |
Use Case Description(s) | |
Use Case Description | 1. Provenance 2. Value-Based Care / ACO Assignment and Attribution 3. Portability of Care Provenance -- critical for other use cases such as security, privacy, and mission-critical clinical needs (e.g. verifying that a vaccine was actually performed or a laboratory test is authentic) -- relies upon organizational identity being unambiguous and, ideally, enumerated in a verifiable way. Value-Based Care / ACO Assignment and Attribution -- while often lumped in with "Coverage" and insurance concepts, an attributable population is often best described as an Organization -- or Group. These organizations are difficult to find in the data, but affect a large number of downstream use cases related to clinical quality reporting, risk-sharing, patient experience measurement and aggregation, and other value-based care activities. Ensuring that the organizations are unique and disambiguated and well-identified is critical to ensuring that patient data is easy to organize and use by ACOs and other stakeholders for value-based care. Portability of Care -- while organizational identifiers have largely focused on NPI-based 'organizations' i.e. provider-assigned identifiers for patients, payors and other non-provider organizations often enumerate individual and group identifiers that apply across different providers and different care settings. So while it may not seem apparent at first, an organizational identifier -- tied to the actual individual identifier granted by that organization -- results in a unique, portable identifier that makes it easier for individuals to transition care between settings, yet ensure continuity between those settings. This results in better portability. |
Estimate the breadth of applicability of the use case(s) for this data element | 300M+ insured Americans with a health insurance identifier likely to be consistent yet portable across healthcare providers. However, every hospital and insurer already captures and exchanges this data element -- they are described elsewhere, just not yet supported in FHIR from an organizational identifier component standpoint. More elaboration and implementation guidance on how to map these organization identifier components into FHIR would begin to introduce it into the standard FHIR Core for software, services and health data infrastructure vendors supporting this community. |
Link to use case project page | https://www.hl7.org/fhir/organization-definitions.html#Organization.identifier |
Healthcare Aims |
|
Maturity of Use and Technical Specifications for Data Element | |
Applicable Standard(s) | For organizations like payers and ACOs, while health plan ID is no longer being pursued by industry, existing, in-production Payer IDs exist in multiple EDI (Electronic Data Interchange) environments. There are a set of Payer IDs in use today already; in concert with a clear description of the system that enumerates the Payer ID, these can be an example of the standards that may be used in uniquely identifying organizations beyond a free-text string description/label. These IDs may, however, be proprietary, but also already permissioned for use by the trading partners of the entities using the IDs for treatment, payment, and health care operations. |
Additional Specifications | N/A |
Current Use | Extensively used in production environments |
Supporting Artifacts |
In original public comment and feedback for prior Health Plan ID efforts, CAQH CORE and others commented on existing payer identifiers (as an example of well-enumerated organization identifiers). The URL link here describes, on page 5, multiple ways of identifying "payers" in existing transactions including payer IDs or other industry-recognized identifiers, such as NAIC numbers, in the HIPAA-mandated X12 standards (TR3) which requires to identify themselves as they choose. While this artifact notes that there is not one central recognized database covering all HIPAA-covered health plans, there are still a finite set of nationally-recognized databases/identifiers to identify a large portion of HIPAA-covered health plans, e.g. NIAC, TIN, or other market-based databases/dictionaries of (proprietary) Payer IDs. https://www.caqh.org/sites/default/files/core/events/2015/q3/07%2001%2015%20HPID%20RFI%20Call%20Deck%20FINAL_0.pdf?token=jYbLil0_ |
Extent of exchange | 5 or more. This data element has been tested at scale between multiple different production environments to support the majority of anticipated stakeholders. |
Supporting Artifacts |
In original public comment and feedback for prior Health Plan ID efforts, CAQH CORE and others commented on existing payer identifiers (as an example of well-enumerated organization identifiers). The URL link here describes, on page 5, multiple ways of identifying "payers" in existing transactions including payer IDs or other industry-recognized identifiers, such as NAIC numbers, in the HIPAA-mandated X12 standards (TR3) which requires to identify themselves as they choose. While this artifact notes that there is not one central recognized database covering all HIPAA-covered health plans, there are still a finite set of nationally-recognized databases/identifiers to identify a large portion of HIPAA-covered health plans, e.g. NIAC, TIN, or other market-based databases/dictionaries of (proprietary) Payer IDs. https://www.caqh.org/sites/default/files/core/events/2015/q3/07%2001%2015%20HPID%20RFI%20Call%20Deck%20FINAL_0.pdf?token=jYbLil0_ |
Potential Challenges | |
Restrictions on Standardization (e.g. proprietary code) | Payer identifiers are sometimes proprietary to the platforms that mediate the electronic transactions; still others are public e.g. Medicare and Medicaid payer identifiers. In general, this code will most likely be acceptable to use for patient care (as it is used for payment already), however the lookup of organizational identifiers may need to be considered for private, rather than public, organizations that maintain and enumerate the payer identifiers. Additionally, private organizations may generate unique, proprietary but free codes for patients to identify themselves, to preserve anonymity but to permit cross-linkage between their medical records. These codes themselves may further not be consistent across provider or payor organizations, but the patient themselves may hold a private master index of all of their unique codes, much as an individual knows all their usernames they use on different systems but does not need to use the same one on every system. |
Restrictions on Use (e.g. licensing, user fees) | Restrictions may be segment-specific, but may also be handled by pre-existing trading agreements. |
Privacy and Security Concerns | Cross-organization identifiers should be treated with general considerations of privacy, security and other attendant focus as any personally-identifiable information. Care should be taken to avoid using an identifier as an authenticator -- the two functions are often confused (e.g. social security number) and the identifier should generally be encouraged not to be an authentication mechanism or token itself. |
Estimate of Overall Burden | Limited or nominal burden to implement in clinical and administrative processes, as the information is already gathered. Unknown burden for software developers and system administrators. In the near-term, so long as the requirement is not Mandatory, the burden will be only nominal. If implemented more broadly, the burden will be upon FHIR-based systems (mostly EHRs and other software applications) and other interfaced systems with the existing identifiers to be able to support the additional identifier sub-elements such as assigner, and also organization identifier sub-elements such as assigner and system. |
Other Implementation Challenges | The mental model of how we identify organizations is not always immediately relevant to implementers, as the focus has often been single-institution data, rather than cross-organization data between not only different providers, but identifiers assigned as foreign keys by non-provider entities e.g. patient-controlled identifier assigners (much like consumer app-generated email aliases) or more commonly, Health Insurers. |
Data Element |
---|
Organizational Identifier Components
Description
Unique, disambiguating identifier components for Organization-related elements. e.g. Provenance, identifier.assigner, and other elements and components that rely upon Organizations, especially those that may not be provider organizations nor have a URL. USCDI may need to support (or even consider requiring) organization.identifier.system to improve data portability. The parent way of identifying the system that enumerates multiple Organization identifiers may best be modeled as Organization.identifier.system, where the identifier uses existing FHIR properties for Identifiers data type. These include: identifier.system (URI namepsace) identifier.assigner (itself an organization!) identifier.value (a unique value) |
|
Submitted by ondec@google.com on 2020-11-18
Example of Organization Identifier: GLEIF
Note heard through Ryan Howells of the CARIN Alliance:
GLEIF is an example of an organizational identifier system that may be generalizable for many use cases, that isn't bound to NPI only. This may be able to help clearly identify other non-provider organizations like health insurers, pharmacy benefits managers, social and civic organizations, and other groups.
For more information about GLEIF:
https://www.gleif.org/en/about-lei/introducing-the-legal-entity-identifier-lei
Henry Wei, MD