LOINC

LOINC (Logical Observation Identifiers Names and Codes) is a universal standard maintained by the Regenstrief Institute for identifying laboratory tests, clinical observations, vital signs, survey instruments, and other clinical measurements. Where SNOMED CT defines what a clinical concept means, LOINC identifies the specific test or observation being ordered or reported. A LOINC code encodes six dimensions: the component being measured, the property measured, the time aspect, the system (specimen type), the scale of measurement, and the method. This multi-axis structure enables precise disambiguation across thousands of test variations.

LOINC is a required standard under the ONC Health IT Certification Program for laboratory result reporting and is referenced throughout the HL7 FHIR specification as the preferred code system for Observation and DiagnosticReport resources. Clinical laboratories, health systems, health information exchanges (HIEs), and public health agencies all rely on LOINC to ensure that a "serum potassium" result from one laboratory system is recognized as semantically equivalent to the same result from another. This semantic consistency is essential for care coordination, population health analytics, and clinical research data aggregation.

LOINC maintains a database of over 100,000 terms covering chemistry, hematology, microbiology, serology, toxicology, radiology reports, vital signs, and patient-reported outcomes. LOINC also defines panels (groupings of individual observations) and answer lists (standardized coded responses for categorical observations). The RELMA mapping tool and the LOINC FHIR API help organizations map their local laboratory codes to LOINC during interface build and system implementation projects.

From an integration engineering perspective, LOINC mapping is required for every laboratory interface, ADT/ORU feed, and clinical data repository that needs to support FHIR, HL7 v2, or C-CDA exchange. Mapping work is labor-intensive: it requires clinical subject matter expertise to correctly assign the six LOINC axes, and many local lab codes require context from the ordering system to be mapped unambiguously. Ongoing maintenance is also required as LOINC releases new terms twice per year and as the laboratory's test menu changes. Automated LOINC mapping tools using NLP and machine learning can accelerate initial mapping but still require expert validation for high-risk or ambiguous codes.

Compliance-Native Architecture Guide

Design principles and a structured checklist for building software that is compliant by default — not compliant by retrofit. Covers data architecture, access controls, audit trails, and vendor due diligence.