Skip to content

Location Profile

Canonical../StructureDefinition/nexus-emr-core-location
Statusdraft · 1.26.0
BaseLocation (constraint)
SourceFSH · JSON

Generic Nexus EMR core profile for Location, represents an EMR 'location' or 'site' -- basically whatever the EMR uses to break the EMR up into separate slices (often for multisite clinics using the same EMR instance) For single-site clinics, at least one Location object must exist that represents the main site: this makes downstream consumers of the FHIR data easier to write as they can always assume Locations exist. Also, note that NexusEmrAppointment REQUIRES a Location as a participant.

Access

What the endpoint offers

Declared for Location by the capability statements this IG publishes. One row per statement; follow it for the rest of what that statement says about this type.

Capability statement Interactions Search parameters
Server Capability Statement read vread history-instance search-type none declared

A cell reading none declared means that statement declares none, which is not the same as the endpoint having none. See how to read a capability statement for what a declared interaction and a declared search parameter do -- and do not -- promise.

SMART on FHIR scopes

Location is reachable by a SMART on FHIR app. These are the scopes an app can request, and the platform permission each one resolves to.

Request this scope Grants Permission required
user/Location.rs Read and search location:read

Location is not in the FHIR patient compartment, so only the user context applies. A patient/Location.rs scope has nothing to narrow to and is not grantable.

Narrowed exposure

Location declares no write or delete permission, so an app cannot create or update or delete it however the scope is written.

Reading the suffix

A clinical scope is <context>/<ResourceType>.<interactions>.

  • Context is patient (the launch patient's data only, and only for resource types in the FHIR patient compartment) or user (everything the authenticated user may see, valid for any type). There is no system context.
  • Interactions are the letters c create, r read, u update, d delete, s search, written in cruds order. Order does not matter on the way in: Location.sr and Location.rs are the same scope.
  • The older SMART v1 suffixes still work and fold on arrival: .read becomes .rs, .write becomes .cud, and .* becomes .cruds.
  • Wildcards (patient/*.rs) expand over the SMART-exposed resource types only, never over everything this IG profiles.

Ask for a whole permission, or none of it

The platform has one read permission covering both r and s, and one write permission covering both c and u. A scope asking for half of either could only be granted by handing over more than it asked for, so it is refused rather than quietly widened: request Location.rs, not Location.r.

An unrecognised or ungrantable scope is rejected outright. It never silently resolves to an empty grant, which a caller could mistake for "no narrowing needed".