Build patient cohorts that mean what you think

Cohort building converts natural language criteria into FHIR searches tagged as inclusion or exclusion concepts, then runs set operations on the patient IDs those searches return.

[
  {
    resource_type: "Patient",
    search_params: "gender=female&birthdate=le1959-01-01",
    exclude: false
  },
  {
    resource_type: "Condition",
    search_params: "code=44054006",
    concept: "diabetes",
    exclude: false
  },
  {
    resource_type: "Condition",
    search_params: "code=38341003",
    concept: "hypertension",
    exclude: true
  }
]

Set operations

• Inclusion concepts are intersected.
• Exclusion concepts are subtracted.
• Patient IDs are pulled from the resource shape, such as Patient.id or subject.reference.

See the Tools reference.