The FHIR Observation resource is one of the most frequently used resources in healthcare interoperability. It represents a single measurement, test result, or clinical assertion about a patient. In laboratory medicine, each individual test result — a hemoglobin value, a glucose level, a white blood cell count — is represented as a separate Observation resource. This granular approach allows each result to carry its own code, value, unit, reference range, and interpretation.
An Observation resource is identified by a code, typically a LOINC code, that specifies exactly what was measured. The result is expressed as a value with a UCUM-coded unit (e.g., 14.2 g/dL for hemoglobin), and the resource includes reference ranges that indicate what is considered normal for that test. An interpretation field can flag results as high, low, critical, or abnormal, providing immediate clinical context. The resource also timestamps when the observation was made and links to the specimen and method used.
One of the strengths of the Observation resource is its ability to represent different types of results. Quantitative results use numeric values with units, qualitative results use coded values (positive, negative, reactive), and semi-quantitative results can use ordinal scales. This flexibility means that virtually any lab test result — from a simple glucose measurement to a complex immunology panel — can be faithfully represented in FHIR.
In lab report digitization, creating accurate Observation resources is the core objective. Each test extracted from a lab report must be mapped to the correct LOINC code, its numeric value parsed and validated, its unit normalized to UCUM, and its reference range captured. The quality of these Observations directly determines the clinical utility of the digitized data — an incorrectly coded or valued Observation could lead to missed diagnoses or inappropriate clinical decisions.