Inferensys

Glossary

LOINC

A universal code system for identifying health measurements, observations, and documents, providing a common language for exchanging laboratory results and clinical assessment scales between disparate systems.
Stylish WeWork-like workspace with hot desks and document wall, professional searching through enterprise knowledge base on a mounted ultrawide display, warm industrial pendants overhead.
UNIVERSAL CLINICAL TERMINOLOGY

What is LOINC?

LOINC (Logical Observation Identifiers Names and Codes) is a universal code system for identifying health measurements, observations, and documents, providing a common language for exchanging laboratory results and clinical assessment scales between disparate systems.

LOINC is a standardized terminology that assigns a unique, permanent identifier to each clinical observation concept, such as a specific laboratory test, vital sign measurement, or standardized survey instrument. Developed by the Regenstrief Institute, it enables semantic interoperability by ensuring that a serum sodium result from one hospital’s EHR carries the same unambiguous meaning when transmitted to a different system or health information exchange.

Each LOINC code represents a fully specified atomic concept defined by six axes, including the component measured, property, timing, system, scale, and method. This granular structure allows mapping to other terminologies like SNOMED CT and is mandated in U.S. Meaningful Use regulations for encoding laboratory orders and results within Consolidated CDA (C-CDA) documents.

UNIVERSAL CODING SYSTEM

Core Characteristics of LOINC

LOINC provides a standardized nomenclature for identifying health measurements, observations, and documents, enabling unambiguous exchange of clinical data between disparate systems.

01

Six-Part Fully Specified Name

Every LOINC term is defined by a unique, structured formal name with six distinct axes that precisely identify a clinical observation:

  • Component: The substance or entity being measured (e.g., Glucose, Systolic Blood Pressure)
  • Property: The characteristic being measured (e.g., Mass Concentration, Pressure)
  • Timing: The temporal aspect of the measurement (e.g., Point in time, 24 Hour)
  • System: The specimen or context (e.g., Serum, Urine, Patient)
  • Scale: The type of result (e.g., Quantitative, Ordinal, Narrative)
  • Method: The procedure used (e.g., Test Strip, Automated Count)

This systematic decomposition ensures that each code represents a single, unambiguous clinical concept, eliminating the ambiguity of local test names like 'Lytes' or 'Chem-7'.

02

Dual-Axis Coverage: Laboratory and Clinical

LOINC is organized into two primary divisions to cover the full spectrum of patient observations:

Laboratory LOINC:

  • Covers chemistry, hematology, microbiology, serology, and toxicology
  • Includes mass spectrometry panels and genetic variant reporting
  • Standardizes the exchange of quantitative lab results between reference labs and EHRs

Clinical LOINC:

  • Encodes vital signs, hemodynamics, and fluid intake/output
  • Standardizes clinical assessment scales (e.g., Glasgow Coma Score, APGAR)
  • Covers radiology reports, discharge summaries, and operative notes as document codes
  • Includes survey instruments like PHQ-9 depression screening

This dual structure allows a single code system to unify both structured lab data and narrative clinical documents under one framework.

03

Document Ontology for Clinical Notes

Beyond lab tests, LOINC provides a robust document ontology to classify clinical notes and reports. This axis uses five attributes to uniquely identify a document type:

  • Type of Service: The kind of clinical activity (e.g., Consultation, Procedure)
  • Subject Matter Domain: The specialty or discipline (e.g., Cardiology, Nursing)
  • Role: The author's function (e.g., Attending Physician, Physical Therapist)
  • Setting: The care environment (e.g., Inpatient, Emergency Department)
  • Kind of Document: The document category (e.g., Progress Note, Discharge Summary)

For example, LOINC code 18842-5 precisely identifies a 'Discharge summary' authored by an attending physician in a cardiology setting, enabling automated document routing and retrieval in health information exchanges.

04

Relationship to Other Terminologies

LOINC does not operate in isolation; it is designed to complement other major clinical terminologies in a coordinated semantic ecosystem:

  • SNOMED CT: Provides the clinical semantics for the Component axis. LOINC codes are mapped to SNOMED CT concepts to link the observation identifier to the clinical meaning of what was measured.
  • UCUM (Unified Code for Units of Measure): LOINC terms reference UCUM codes to specify the exact units of measure for quantitative results (e.g., mg/dL, mmol/L), ensuring computational comparability.
  • HL7 v2 and FHIR: LOINC is the required code system for the OBX-3 field in HL7 v2 messages and the Observation.code element in FHIR resources, making it the universal identifier for test results in all modern interoperability standards.
  • ICD-10-CM: While ICD codes classify diagnoses, LOINC codes identify the observations that support those diagnoses, creating a linked chain of clinical evidence.
05

Common LOINC Panels and Examples

LOINC groups related individual tests into panels to represent standard order sets and their expected results:

  • Basic Metabolic Panel (BMP): LOINC 24323-8 bundles codes for Glucose (2345-7), Sodium (2951-2), Potassium (2823-3), Chloride (2075-0), CO2 (2028-9), BUN (3094-0), Creatinine (2160-0), and Calcium (17861-6).
  • Complete Blood Count (CBC): LOINC 58410-2 organizes White Blood Cell Count (6690-2), Red Blood Cell Count (789-8), Hemoglobin (718-7), Hematocrit (4544-3), and Platelet Count (777-3).
  • Urinalysis Panel: LOINC 24356-8 includes pH (2756-5), Protein (20454-5), Glucose (25428-4), Ketones (2514-8), and Nitrite (5802-4).

Using panel codes allows systems to transmit a single order identifier while the receiving system expands it into the expected child observations, reducing message complexity.

06

Versioning and Release Cycle

LOINC is actively maintained by the Regenstrief Institute with a rigorous, predictable release cycle to keep pace with evolving clinical practice:

  • Semi-annual releases: New versions published each February and August
  • Current scope: Over 98,000 active terms covering laboratory, clinical, and document concepts
  • Deprecation policy: Codes are marked as deprecated rather than deleted, ensuring historical data remains interpretable
  • RELMA mapping tool: A freely available Windows application that assists institutions in mapping their local test catalogs to LOINC codes using lexical matching and synonymy
  • Community submissions: New term requests are reviewed by clinical experts and added based on evidence of widespread use

This governance model ensures LOINC remains a living standard that adapts to new biomarkers, genomic tests, and clinical assessment instruments without breaking backward compatibility.

LOINC CLARIFIED

Frequently Asked Questions

Clear, technically precise answers to the most common questions about the Logical Observation Identifiers Names and Codes (LOINC) standard, its structure, and its role in clinical data interoperability.

LOINC (Logical Observation Identifiers Names and Codes) is a universal code system developed by the Regenstrief Institute that provides a definitive, unambiguous identifier for health measurements, observations, and clinical documents. It functions by assigning a unique, permanent code to each distinct test or observation, defined by a formal six-axis model: Component (what is measured), Property (the characteristic, like mass concentration), Time (point or interval), System (the specimen or context), Scale (quantitative, ordinal, nominal), and Method (the procedure). This structured nomenclature allows a serum sodium test from a Quest Diagnostics analyzer and one from a LabCorp instrument to be identified with the identical LOINC code, enabling semantically consistent aggregation and exchange across disparate electronic health record (EHR) systems.

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.