Glossary

FHIR Backend Glossary

A working glossary of the FHIR backend concepts that show up in product, engineering, security, and clinical-informatics conversations. Each term has a one-line definition for skimming and a dedicated page for context, common pitfalls, and how Fire Arrow handles it.

If you only read three

Start with FHIR backend: it sets the scope of what a healthcare application backend has to cover beyond pure FHIR storage. Then read FHIR RBAC: the authorisation model that healthcare actually needs (compartments, organisations, care teams, identity-conditional rules) instead of a generic role table. Finally, read FHIR Subscription: the workflow primitive that connects the backend to whatever the application does next, from patient reminders to clinical alerts to downstream analytics.

Foundations

FHIR backend
A FHIR backend is a server that uses FHIR resources as the application data model and exposes them through FHIR APIs (REST, GraphQL) along with the application services around them: authentication, authorization, scheduling, notifications, and admin tooling.
FHIR server
A FHIR server is an HTTP service that implements the FHIR REST API: read, search, create, update, delete, history, and the standard operations defined by HL7 FHIR.
FHIR resource
A FHIR resource is a typed clinical data object (Patient, Observation, CarePlan, MedicationRequest, QuestionnaireResponse, Task, and others) defined by the HL7 FHIR specification.
HAPI FHIR
HAPI FHIR is an open-source Java framework for FHIR. It includes a JPA-based server implementation widely used as a foundation for production FHIR backends.

API surfaces

FHIR GraphQL
FHIR GraphQL is a GraphQL query interface over FHIR resources. Clients request the fields and related resources they need in one query, instead of multiple FHIR REST round trips.
SMART on FHIR
SMART on FHIR is a profile that defines OAuth 2.0 launch flows and scope strings for applications that consume FHIR data, typically launched from an EHR.

Authorisation and privacy

FHIR RBAC
FHIR RBAC is role-based access control applied to FHIR resources, operations, and the relationships between them, rather than to abstract resource types alone.
FHIR compartment
A FHIR compartment is a logical grouping of resources related to a compartment owner (typically a Patient, Practitioner, RelatedPerson, Device, or Encounter). It is defined in the FHIR specification per resource type.
FHIR de-identification
FHIR de-identification is the practice of removing or transforming identifying fields from FHIR resources so that the resulting data can be used for purposes that should not have access to direct patient identifiers.

Workflow primitives

CarePlan in FHIR
A CarePlan is a FHIR resource that describes intended care for a patient: activities, goals, care team, status, and timeline.
Task in FHIR
A Task is a FHIR resource that represents a discrete unit of work: who should do it, what it is for, what its current status is, and when it is due or was completed.
QuestionnaireResponse
A QuestionnaireResponse is a FHIR resource that holds the answers a patient or clinician submitted to a Questionnaire, with structured items linked back to the Questionnaire definition.
FHIR Subscription
A FHIR Subscription is a resource that tells a FHIR server to deliver a notification when resources matching a search expression are created or updated.

Profiles and implementation guides

FHIR implementation guide
A FHIR implementation guide (IG) is a published bundle of profiles, value sets, code systems, examples, and narrative documentation that describes how to use FHIR for a specific purpose.
US Core
US Core is the HL7 FHIR implementation guide that defines the minimum profiles and value sets US healthcare systems use for interoperability, including USCDI conformance.
ISiK profile
ISiK (Informationstechnische Systeme im Krankenhaus) is the FHIR profile family the gematik defines for the German hospital information system ecosystem and its data-exchange partners.
MII FHIR profile
The MII (Medizininformatik-Initiative) profiles are the FHIR profiles defined by the German Medical Informatics Initiative for cross-site research data sharing across university hospitals.
International Patient Summary (IPS)
The International Patient Summary (IPS) is an HL7 FHIR profile defining a portable, condensed clinical summary intended for use across borders and across health systems.
FHIR validation
FHIR validation checks resources against the FHIR base specification and any applied profiles for structural correctness, value-set conformance, and required-element presence.

FHIR versions

FHIR R5
FHIR R5 is the fifth major release of the HL7 FHIR specification, published in 2023, with normative additions to several resource types and changes to terminology and search behaviour.
FHIR R4 vs R5
FHIR R4 (2019) is the version most production deployments and regulators reference; FHIR R5 (2023) adds new resource types, refined search semantics, and the topic-based subscription model.

Audit and consent

FHIR Provenance
FHIR Provenance is the resource type that captures the origin and history of other FHIR resources: who created or modified them, when, with what role, and on the basis of what source.
FHIR AuditEvent
FHIR AuditEvent is the resource type that records access to and actions on FHIR resources, supporting regulatory audit and security investigation.
FHIR Consent
FHIR Consent is the resource type that records a patient's choices about how their data may be accessed, used, or disclosed, including provisions and exceptions.

Bulk and CDS

FHIR Bulk Data
FHIR Bulk Data is the FHIR specification for asynchronous, large-scale dataset export from a FHIR server, using the $export operation and NDJSON output.
CDS Hooks
CDS Hooks is a HL7 specification for triggering external clinical decision support during specific points in clinician workflows (opening a chart, ordering a medication, signing a note) over a standard webhook protocol.

AI

MCP for healthcare
Model Context Protocol (MCP) for healthcare is the application of MCP, a standard interface for LLM agents to call external tools, to clinical systems where the tools include FHIR reads, writes, and operations.