HL7 FHIR Interoperability

When to use

• Implementing FHIR R4 CRUD or search operations against any server
• Standing up a HAPI FHIR or Microsoft FHIR Server instance
• Building a SMART on FHIR application (EHR launch or standalone)
• Implementing a CDS Hooks service for point-of-care decision support
• Running a bulk data export for population health or quality reporting
• Validating resources against US Core Implementation Guide profiles

Core principles

1. Capability statement is your contract — publish /metadata accurately; clients depend on it to know what you actually support
2. Search parameters must be indexed — unindexed searches on large datasets will time out in production; know your server's indexing model
3. Conditional create prevents duplicates — use If-None-Exist headers on POST; without them, retries create duplicate resources
4. SMART scopes are minimum necessary — request only the scopes the app actually needs; over-scoped tokens are a liability
5. Bulk export is async by design — poll the Content-Location URL; do not block a thread waiting for $export to complete

Reference Files

• references/fhir-resources-api.md — core resource table, CRUD operations, search syntax with system/code, _include/_revinclude, Bundle pagination, $everything/$validate/$export operations
• references/fhir-server-implementation.md — HAPI FHIR setup and interceptors, Microsoft FHIR Server with Azure AD, versioning, conditional create/update, Subscription patterns, capability statement
• references/smart-cds-bulk.md — SMART on FHIR authorization flow, OAuth scopes, EHR and standalone launch contexts, CDS Hooks hook points and card format, Bulk Data NDJSON export, US Core profiles and USCDI alignment