REST, or Representational State Transfer, is an architectural style for designing distributed systems and web services. Defined by Roy Fielding in 2000, REST provides a set of principles that guide how resources are identified, addressed, and manipulated over a network. RESTful systems use standard HTTP methods — GET to retrieve data, POST to create resources, PUT to update them, and DELETE to remove them — making them intuitive for developers familiar with web technologies.
The REST architectural style is built on several key principles. Resources are identified by URIs (Uniform Resource Identifiers), and each resource can have multiple representations (JSON, XML, HTML). Interactions are stateless, meaning each request contains all the information needed to process it. Responses can be cached to improve performance. These principles result in systems that are scalable, maintainable, and loosely coupled — qualities that are essential for healthcare data exchange.
HL7 chose REST as the foundation for FHIR specifically because of its simplicity and widespread adoption. In a FHIR RESTful API, clinical resources like Patients, Observations, and DiagnosticReports are accessed through predictable URL patterns. For example, GET /Observation?patient=123&code=718-7 retrieves all hemoglobin observations for a specific patient. This consistency makes FHIR APIs accessible to general-purpose developers, not just healthcare IT specialists, dramatically expanding the pool of talent that can build health applications.
Lab report digitization APIs typically follow RESTful conventions. A client submits a lab report image via a POST request to a processing endpoint and receives the structured FHIR data in the response. The API may support additional RESTful patterns like polling for asynchronous results, paginating large result sets, or content negotiation to return data in different formats. By adhering to REST principles, lab processing APIs integrate seamlessly into the broader ecosystem of web services and healthcare applications.