Shorts on Standards: Regenstrief Institute and SNOMED International release 1st version of LOINC ontology

A LOINC and SNOMED CT Interoperability Solution under a new collaborative agreement

The CAP has dozens of liaisons to various organizations and efforts, and periodically we publish briefs of what the liaisons hear and see in their liaison roles. Our latest, below, is from the CAP Informatics Committee.

Hung S. Luu, MD, PharmD; Simone Arvisais-Anhalt, MD; Rajesh C. Dash, MD

June 2025—Laboratories are subject to rigorous federal regulations such as continuous quality control, proficiency testing, and laboratory inspections to ensure high-quality test performance. However, few government mandates exist for the standardization of laboratory data. As a result, different laboratories have various local codes, names, reference ranges for normal and abnormal values, and formats of results and associated units for laboratory tests that might otherwise be similar or identical. While the lack of standardization is unlikely to impact the daily operations of individual laboratories, it does pose a significant barrier when results are compared or aggregated across institutions. It also poses challenges for secondary uses such as public health surveillance, outcomes analysis, research, and artificial intelligence/machine learning development.

Two standard terminologies used for laboratory data are the Logical Observation Identifiers Names and Codes (LOINC) and Systematized Nomenclature of Medicine–Clinical Terms (SNOMED CT). These were developed to facilitate the exchange of health care data and address the need for unambiguous clinical data representation. Laboratory tests and results can be thought of as a question-answer pair where the questions asked (i.e. the lab test performed) were intended to be represented by LOINC terms, while the answer (i.e. qualitative result) would be described with SNOMED CT codes.

The Regenstrief Institute in 1994 developed LOINC to provide universal identifiers for the OBX-3 field of the Health Level 7 observation-reporting message, thereby eliminating confusion regarding local naming conventions and representation of test orders and results. The LOINC terminology uses numeric codes that identify the type of laboratory test with defined attributes based on six axes: <Component or analyte>: <Kind of property>: <Time aspect>: <Sample type>: <Scale of measurements>: <Method>. Criticisms of LOINC have included lack of a hierarchical structure that allows for data aggregation of related granular concepts under a more general structure and lack of a computable infrastructure to enable logical inferences and relationships between parts, terms, and concepts.

A cooperative agreement was executed between the Regenstrief Institute and SNOMED International in 2022 and renewed this year. A primary goal of the agreement is to create an extension of SNOMED CT for the LOINC terminology to minimize duplication, enhance stakeholders’ ability to use the terminologies together, and provide laboratory and pathology content in a recognized format to countries that do not currently use LOINC. The agreement is intended to strengthen the two terminologies by increasing the computability of LOINC and making the more granular observables in LOINC available to be used in a SNOMED format integrated with SNOMED CT concepts.

In March this year, version 1.0 of the LOINC Ontology: A LOINC and SNOMED CT Interoperability Solution was published. The LOINC Ontology is an extension of SNOMED CT’s ontological framework and creates SNOMED CT and LOINC codes for all concepts shared between the terminologies. The first release includes more than 35,000 laboratory tests, including the 200 most used tests as defined by the Regenstrief Institute using data sets from Observational Health Data Sciences and Informatics and the National Patient-Centered Clinical Research Network.

Dr. Luu is professor of pathology at UT Southwestern Medical Center and director of clinical pathology at Children’s Health, a pediatric health care system. He is chair of the CAP Informatics Committee and the CAP’s liaison to the LOINC effort. Dr. Arvisais-Anhalt is assistant professor of laboratory medicine at the University of California San Francisco and director of laboratory medicine informatics. She is a member of the CAP Informatics Committee. Dr. Dash is professor of pathology, Duke University Health System, and vice chair for anatomic pathology and informatics. He is a member of the Informatics Committee and former chair of the CAP Artificial Intelligence Committee. He is planning co-chair for the Integrating the Healthcare Enterprise, Pathology and Laboratory Medicine (IHE PaLM) domain, focused on lab data interoperability.