Large Language Models' Performances regarding logical observation identifiers names and codes mapping in laboratory medicine: A comparative analysis of ChatGPT-4.0, Gemini, and Perplexity.

ObjectiveTo examine the use of LOINC and SNOMED CT codes for coding laboratory orders and results in laboratory reports sent from 63 non-federal hospitals to the BioSense Program in calendar year 2011.IntroductionMonitoring laboratory test reports could aid disease surveillance by adding diagnostic specificity to early warning signals and thus improving the efficiency of public health investigation of detected signals. Laboratory data could also be employed to direct and evaluate interventions and countermeasures, while monitoring outbreak trends and progress; this would ultimately result in better outbreak response and management, and enhanced situation awareness. Since Electronic Laboratory Reporting (ELR) has the potential to be more accurate, timely, and cost-effective than reporting by other means of communication (e.g., mail, fax, etc.), ELR adoption has been systematically promoted as a public health priority. However, the continuing use of non-standard, local codes or text to represent laboratory test type and results complicates the use of ELR data in public health practice. Use of structured, unique, and widely available coding system(s) to support the concepts represented by locally assigned laboratory test order and result information improves the computational characteristics of ELR data. Out of several coding strategies available, the Office of the U.S. National Coordinator for Health Information Technology has recently suggested incorporating Logical Observation Identifiers Names and Codes (LOINC) for laboratory orders and Systemized Nomenclature of Medicine- Clinical Terms (SNOMED CT) codes for laboratory results to standardize ELR.MethodsWe assessed the use of LOINC and SNOMED CT codes in laboratory data reported to BioSense, a near real-time national-level, electronic syndromic surveillance system, managed by the Centers for Disease Control and Prevention. ELR data reported by 63 non-federal hospitals to BioSense in 2011 were analyzed to examine LOINC and SNOMED CT use in coding laboratory orders and results. We used Relma software, developed and distributed by Regenstrief Institute Inc for identifying LOINC codes.ResultsIn 2011, a total of 14,028,774 laboratory test order or result reports from 821,108 individual patients were reported from the 63 hospitals in 14 states. Since, by design the BioSense Program monitors a select set of syndromes mainly representing infectious conditions, 94% of the total reports were microbiology test orders or results. Seventy-seven percent of all test orders (n = 10,776,494) used LOINC codes. Of all test results with at least one value either in observation identifier (OBX3) or observation value (OBX5) segments of their Health Level 7 (HL7) ELR message (n = 12,313,952), 81% had only LOINC codes, 0.1% had only SNOMED codes, 7% had both LOINC and SNOMED codes, and 12% used no codes. In total, 1,428 unique LOINC and 608 unique SNOMED codes were used to describe the results, and 805 unique LOINC codes were used to describe the orders. Of the 608 unique SNOMED codes, 111 (18.3%) did not have corresponding LOINC codes. Fifty-one (46%) of these 111 SNOMED codes could have been matched to corresponding LOINC codes based on the concept. However, our search for matching LOINC codes in Relma for certain SNOMED concepts indicated that LOINC does not have codes for select types of laboratory test results, particularly qualifier (such as reactive, negative, and resistant) or structural (labia, urethra, and vagina) concepts.ConclusionsOur analysis showed that the use of SNOMED CT codes for laboratory test results by non-federal hospitals reporting laboratory data to BioSense was extremely limited. These hospitals more frequently used LOINC codes than SNOMED CT in reporting test results. We found that a large percentage of test results with SNOMED CT codes could be represented by LOINC codes that exactly or closely match SNOMED CT codes. Using LOINC codes to report both test order and results in these databases could increase the availability and use of laboratory data in public health and surveillance activities. However, to increase the sensitivity of the coding further, a small number of tests could benefit by using LOINC along with SNOMED CT codes. Evaluation of use of syndromic surveillance case definitions that incorporate laboratory result information is required to determine if it improves syndromic surveillance performance for enhanced outbreak detection or improved situation awareness.

ObjectiveThe purpose of this project is to demonstrate the progress in development of a standardized public health (PH) emergency preparedness and response data ontology (terminology) through collaboration between the Centers for Disease Control and Prevention (CDC), Division of Emergency Operations (DEO), and the Logical Observation Identifiers Names and Codes (LOINC) system.IntroductionThe U.S. Department of Homeland Security National Incident Management System (NIMS) establishes a common framework and common terminology that allows diverse incident management and support organizations to work together across a wide variety of functions and hazard scenarios1. Using common terminology helps avoid confusion and enhances interoperability, particularly in fast-moving public health (PH) emergency responses. In addition, common terminology allows diverse incident management and support organizations to work together across a wide variety of functions and scenarios1. LOINC is one of a suite of designated standards for the electronic exchange of public health and clinical information. Implementation of LOINC facilitates improvement of semantic interoperability, including unified terminology2. More than 68,100 registered users from 172 countries use LOINC to move interoperable data seamlessly between systems3. The CDC Division of Emergency Operations (DEO) leads development of standardized PH emergency preparedness and response terminology to improve effective and interoperable communications between national and international partners. Realizing the scale of LOINC support and implementation across the global public health arena, CDC DEO collaborates with LOINC to further enhance and harmonize the current PH emergency response terminology and to attain critical PH emergency management and preparedness and response requirements.MethodsDEO analyzed 87,863 LOINC terms that were included in LOINC version 2.64, released on 06/15/20183. Access to this LOINC version was obtained through the Regenstrief LOINC Mapping Assistant (RELMA). RELMA is a Windows-based LOINC utility developed by the Regenstrief Institute (Indiapolis, USA) for searching the LOINC database and mapping local codes to LOINC codes4. The relevance of LOINC terminology to PH emergency preparedness and response was assessed through evaluating existing LOINC terminology against terminology specified by the World Health Organization PH Emergency Operation Centers (EOC). The following functions were evaluated: 1) Managing and Commanding; 2) Operating; 3) Planning/Intelligence; 4) Logistics and 5) Finance/Administration5. LOINC terminology was also evaluated against the CDC PH EOC Minimum Data Set (MDS)6 that contains 315 standardized terms. Analysis of fully specified LOINC terms was conducted through assessment of such LOINC term parts (attributes) as the code, name (component), system, method and class. Recommendations of gaps and enhancements were coordinated with LOINC management for inclusion of the new terminology in the release of version 2.65 .ResultsA new LOINC method, “CDC.EOC”, is under development. Currently, the “emergency management incident” terminology presented by LOINC is limited by such characteristics as event type, event location and event name and requires amplification regarding to PH operations (i.e., communication, logistics etc.).As a result of this investigation, emergency management terms are now being classified according to the type of incident or event (i.e., hurricane, outbreak, etc.) under LOINC code 80394-0. Similarly laboratory and clinical terms are being classified under a provisional LOINC code (89724-9). Two panels were created: 1) The emergency medical systems from the National Emergency Medical Services Information System (NEMSIS) was added under the NEMSIS.Panel (n= 177 terms) and 2) the Data Elements for Emergency Departments Systems (DEEDS) panel (n = 152 terms) was added with two subpanels: Attach.ED and Panel.ED.Assessing existing LOINC taxonomy and codification, DEO is working with the LOINC management team on evaluating additional options for reconciliation the PH emergency preparedness and response common information exchange reference model and LOINC standard. This process aims to further improve semantic interoperability of PH emergency preparedness and response information.ConclusionsThe LOINC terminology standardization is essential for improving PH preparedness and response data exchange and semantic interoperability. Collaboration with the Regenstrief Institute (LOINC) allows CDC to meet the terminology needs of PH emergency management and defines new opportunities for reconciliation data exchange between NIMS partners. This collaborative effort incorporates critically needed PH emergency and preparedness terminology and hierarchical structure in the LOINC standard.

Logical Observation Identifiers Names and Codes (LOINC) are a standard for identifying and reporting laboratory investigations that were developed and are maintained by the Regenstrief Institute. LOINC codes have been adopted globally by hospitals, government agencies, laboratories, and research institutions. There are still many healthcare organizations, however, that have not adopted LOINC codes, including rural hospitals in low- and middle- income countries. Hence, organizations in these areas do not receive the benefits that accrue with the adoption of LOINC codes. We conducted a literature search by utilizing PubMed, CINAHL, Google Scholar, ACM Digital Library, and the Biomed Central database to look for existing publications on the benefits and challenges of adopting LOINC. We selected and reviewed 16 publications and then conducted a case study via the following steps: (1) we brainstormed, discussed, analyzed, created and revised iteratively the patient's clinical encounter (outpatient or ambulatory settings) process within a laboratory department via utilizing a hypothetical patient; (2) we incorporated the work experience of one of the authors (CU) in a rural hospital laboratory department in Nigeria to break down the clinical encounter process into simpler and discrete steps and created a series of use cases for the process; (3) we then analyzed and summarized the potential usage of LOINC codes (clinically, administratively, and operationally) and the benefits and challenges of adopting LOINC codes in such settings by examining the use cases one by one. Based on the literature review, we noted that LOINC codes' ability to improve laboratory results' interoperability has been recognized broadly. LOINC-coded laboratory results can improve patients' safety due to their consistent meaning as well as the related reduction of duplicate lab tests, easier assessment of workloads in the laboratory departments, and accurate auditing of laboratory accounts. Further, the adoption of LOINC codes may motivate government agencies to upgrade hospitals' infrastructures, which could increase the possibility of international recognition of laboratory test results from those hospitals over the long term. Meanwhile, a lack of LOINC codes in paper format and a lack of LOINC codes experts are major challenges that may limit LOINC adoption. In this paper, we intend to provide a snapshot of the possible usage of LOINC codes in rural hospitals in low- and middle-income countries via simpler and detailed use cases. Our analysis may aid policymakers to gain a deeper understanding of LOINC codes in regard to clinical, administrative, and operational aspect and to make better-informed decisions in regard to LOINC codes adoption. The use case analysis also can be used by information system designers and developers to reference workflow within a laboratory department. We recognize that this manuscript is only a case study and that the exact steps and workflows may vary in different laboratory departments; however, the core steps and main benefits should be consistent.

ObjectiveElectronic laboratory reporting has been promoted as a public health priority.The Office of the U.S. National Coordinator for Health Information Technologyhas endorsed two coding systems: Logical Observation Identifiers Names and Codes(LOINC) for laboratory test orders and Systemized Nomenclature ofMedicine-Clinical Terms (SNOMED CT) for test results.Materials and MethodsWe examined LOINC and SNOMED CT code use in electronic laboratory data reportedin 2011 by 63 non-federal hospitals to BioSense electronic syndromicsurveillance system. We analyzed the frequencies, characteristics, and codeconcepts of test orders and results.ResultsA total of 14,028,774 laboratory test orders or results were reported. No testorders used SNOMED CT codes. To describe test orders, 77% used a LOINC code, 17%had no value, and 6% had a non-informative value, “OTH”.Thirty-three percent (33%) of test results had missing or non-informative codes.For test results with at least one informative value, 91.8% had only LOINCcodes, 0.7% had only SNOMED codes, and 7.4% had both. Of 108 SNOMED CT codesreported without LOINC codes, 45% could be matched to at least one LOINCcode.ConclusionMissing or non-informative codes comprised almost a quarter of laboratory testorders and a third of test results reported to BioSense by non-federalhospitals. Use of LOINC codes for laboratory test results was more common thanuse of SNOMED CT. Complete and standardized coding could improve the usefulnessof laboratory data for public health surveillance and response.

BackgroundThe accuracy of Logical Observation Identifiers Names and Codes (LOINC) mappings is reportedly low, and the LOINC codes used for research purposes in Korea have not been validated for accuracy or usability. Our study aimed to evaluate the discrepancies and similarities in interoperability using existing LOINC mappings in actual patient care settings.MethodsWe collected data on local test codes and their corresponding LOINC mappings from seven university hospitals. Our analysis focused on laboratory tests that are frequently requested, excluding clinical microbiology and molecular tests. Codes from nationwide proficiency tests served as intermediary benchmarks for comparison. A research team, comprising clinical pathologists and terminology experts, utilized the LOINC manual to reach a consensus on determining the most suitable LOINC codes.ResultsA total of 235 LOINC codes were designated as optimal codes for 162 frequent tests. Among these, 51 test items, including 34 urine tests, required multiple optimal LOINC codes, primarily due to unnoted properties such as whether the test was quantitative or qualitative, or differences in measurement units. We analyzed 962 LOINC codes linked to 162 tests across seven institutions, discovering that 792 (82.3%) of these codes were consistent. Inconsistencies were most common in the analyte component (38 inconsistencies, 33.3%), followed by the method (33 inconsistencies, 28.9%), and properties (13 inconsistencies, 11.4%).ConclusionThis study reveals a significant inconsistency rate of over 15% in LOINC mappings utilized for research purposes in university hospitals, underlining the necessity for expert verification to enhance interoperability in real patient care.

BackgroundAmerican Indian and Alaska Native (AI/AN) populations have high rates of Respiratory Syncytial Virus (RSV)-associated hospitalizations among both infants and older populations. Due to underrepresentation, most national surveillance reports do not adequately describe the burden of disease in the AI/AN population. Currently, there is no standardized approach for RSV surveillance within the IHS. This work explores establishing RSV surveillance using IHS-specific electronic medical records (EMR) data. Additionally, this work aims to implement a standardized surveillance methodology across IHS for RSV and adaptable for other infectious diseases. The objective of this abstract is to describe the process of RSV surveillance methodology within the IHS.Venn diagram of ICD and LOINC datasetsThe above image is a representation of the Venn diagram of ICD and LOINC datasets. The two datasets are merged together to identify the overlapped records and unique ICD and LOINC records for RSV.MethodsWe investigated the utility of using administrative data from Federal, Tribal, and Urban Indian health facilities for RSV surveillance among AI/AN population for Fiscal Year (FY) 2023. We evaluated RSV administrative codes, such as Logical Observation Identifiers Names and Codes (LOINC), International Classification of Diseases (ICD) 9 and 10, to develop an algorithm for RSV records. The records that contained either an ICD 9 or ICD 10 code for RSV came from the ICD dataset; records that contained RSV-specific LOINC code came from the LOINC dataset. Final set of records were flagged as positive and negative RSV records.Table 1.Distribution of unique LOINC records by the type of testsThe table shows the distribution of the unique LOINC records by the four categories of type of tests. The categories include PCR, Antigen, Antibody and Culture tests.ResultsIn FY 2023, 16396 records came from the ICD dataset and 151542 records came from the LOINC dataset. On merging ICD and LOINC datasets, 4623 records overlapped, 11830 records were unique to ICD dataset and 146959 records were unique to LOINC dataset. Of the records unique to LOINC dataset (146959), 81% (n=119698) represent PCR tests while antigen (n=13722) and antibody (n=13384) tests represent 9% respectively.ConclusionThis study highlights the importance of using both ICD and LOINC codes for identifying RSV cases in EMR. It aids in understanding trends and burden among AI/AN population. Currently, we are conducting site validations to evaluate the surveillance methods across IHS. Additionally, surveillance reports and epidemiological studies generated using this standard approach could help target interventions for RSV and other infectious diseases.DisclosuresAll Authors: No reported disclosures

Semantic interoperability in clinical processes is necessary to exchange meaningful information among healthcare facilities. Standardized classification and coding systems allow for meaningful information exchange. This paper aims to support the accuracy validation of mappings between local and standardized clinical content, through the construction of NooJ syntactical grammars for recognition of local linguistic forms and detection of data correctness level. In particular, this work deals with laboratory observations, which are identified by idiosyncratic codes and names by different facilities, thus creating issues in data exchange. The Logical Observation Identifiers Names and Codes (LOINC) is an international standard for uniquely identifying laboratory and clinical observations. Mapping local concepts to LOINC allows to create links among health data systems, even though it is a cost and time-consuming process. Beyond this, in Italy LOINC experts use to manually double check all the performed mappings to validate them. This has over time become a non-trivial task because of the dimension of laboratory catalogues and the growing adoption of LOINC. The aim of this work is realizing a NooJ grammar system to support LOINC experts in validating mappings between local tests and LOINC codes. We constructed syntactical grammars to recognize local linguistic forms and determine data accuracy, and the NooJ contextual constraints to identify the threshold of correctness of each mapping. The grammars created help LOINC experts in reducing the time required for mappings validation.

LOINC® Codes for Hospital Information Systems Documents: A Case Study

Background and Objective: Logical Observation Identifiers Names and Codes (LOINC) is a universal standard for identifying laboratory tests and clinical observations. It facilitates a smooth information exchange between hospitals, locally and internationally. Although it offers immense benefits for patient care, LOINC coding is complex, resource-intensive, and requires substantial domain expertise. Our objective was to provide training and evaluate the performance of LOINC mapping of 20 pathogens from 53 hospitals participating in the National Notifiable Disease Surveillance System (NNDSS). Methods: Complete mapping codes for 20 pathogens (nine bacteria and 11 viruses) were requested from all participating hospitals to review between January 2014 and December 2016. Participating hospitals mapped those pathogens to LOINC terminology, utilizing the Regenstrief LOINC mapping assistant (RELMA) and reported to the NNDSS, beginning in January 2014. The mapping problems were identified by expert panels that classified frequently asked questionnaires (FAQs) into seven LOINC categories. Finally, proper and meaningful suggestions were provided based on the error pattern in the FAQs. A general meeting was organized if the error pattern proved to be difficult to resolve. If the experts did not conclude the local issue’s error pattern, a request was sent to the LOINC committee for resolution. Results: A total of 53 hospitals participated in our study. Of these, 26 (49.05%) used homegrown and 27 (50.95%) used outsourced LOINC mapping. Hospitals who participated in 2015 had a greater improvement in LOINC mapping than those of 2016 (26.5% vs. 3.9%). Most FAQs were related to notification principles (47%), LOINC system (42%), and LOINC property (26%) in 2014, 2015, and 2016, respectively. Conclusions: The findings of our study show that multiple stage approaches improved LOINC mapping by up to 26.5%.

This paper describes the experience of mapping the document types defined by Medas company, a Digital Preservation Organization for healthcare institutions, to the Logical Observation Identifiers Names and Codes (LOINC) international standard, and consequently assess its adequacy in representing specificities of the Italian context. Mapping operations were manually performed by LOINC Italia experts. The LOINC database was searched using the LOINC Search web browser and REgenstrief LOINC Mapping Assistant (RELMA) software, employing local names and synonyms of Medas document types. Out of 483 Medas document types, 144 fully match with a LOINC code; 211 were associated to a LOINC code which only partially covers the meaning; 128 are not covered by LOINC. Although the axes of LOINC Document Ontology are quite adequate for document types representation, an extension of Kind of document and Subject Matter Domain axes is desirable. Local Medas codes reflects the variety of clinical document types produced in the Italian healthcare domain and the results of this case study can be used as new terms submissions to enrich the LOINC standard, especially in relation to the specificities that characterize the Italian national context.

The Logical Observation Identifiers Names and Codes (LOINC) system is supposed to facilitate interoperability, and it is the federally required code for exchanging laboratory data. To provide an overview of LOINC, emerging issues related to its use, and areas relevant to the pathology laboratory, including the subtleties of test code selection and importance of mapping the correct codes to local test menus. This review is based on peer-reviewed literature, federal regulations, working group reports, the LOINC database (version 2.65), experience using LOINC in the laboratory at several large health care systems, and insight from laboratory information system vendors. The current LOINC database contains more than 55 000 numeric codes specific for laboratory tests. Each record in the LOINC database includes 6 major axes/parts for the unique specification of each individual observation or measurement. Assigning LOINC codes to a laboratory's test menu should be a defined process. In some cases, LOINC can aid in distinguishing laboratory data among different information systems, whereby such benefits are not achievable by relying on the laboratory test name alone. Criticisms of LOINC include the complexity and resource-intensive process of selecting the most correct code for each laboratory test, the real-world experience that these codes are not uniformly assigned across laboratories, and that 2 tests that may have the same appropriately assigned LOINC code may not necessarily have equivalency to permit interoperability of their result data. The coding system's limitations, which subsequently reduce the potential utility of LOINC, are poorly understood outside of the laboratory.

Comparative study of LOINC and SNOMED CT in panel mapping: enhancing interoperability in laboratory testing.

In Hong Kong Hospital Authority (HA), the Electronic Patient Record (ePR) module of Clinical Management System, implemented since 2003, bring together all the information from various clinical module and hospitals into a single corporate wide, longitudinal, integrated record. Nowadays there are billions of laboratory test results stored in the web-based ePR where laboratory results being shared with the HA clinicians for patient care. In order to produce interoperable laboratory data in the ePR, the HA adopts LOINC (Logical Observation Identifiers Names and Codes) as the reference standard for laboratory tests. Every local test codes have been mapped with LOINC code where possible. Thus, the accuracy of LOINC mapping for laboratory tests in the HA is imperative. This paper describes a quality assurance program of LOINC mapping for laboratory tests conducted in 2011/12. With the use of right people, right process and right technology, we reviewed over 28,000 local test codes and there are around 2,400 distinct LOINC concepts mapped and defined in the system.

MotivationMapping internal, locally used lab test codes to standardized logical observation identifiers names and codes (LOINC) terminology has become an essential step in harmonizing electronic health record (EHR) data across different institutions. However, most existing LOINC code mappers are based on text-mining technology and do not provide robust multi-language support.Materials and methodsWe introduce a simple, yet effective tool called big data-guided LOINC code mapper (BGLM), which leverages the large amount of patient data stored in EHR systems to perform LOINC coding mapping. Distinguishing from existing methods, BGLM conducts mapping based on distributional similarity.ResultsWe validated the performance of BGLM with real-world datasets and showed that high mapping precision could be achieved under proper false discovery rate control. In addition, we showed that the mapping results of BGLM could be used to boost the performance of Regenstrief LOINC Mapping Assistant (RELMA), one of the most widely used LOINC code mappers.ConclusionsBGLM paves a new way for LOINC code mapping and therefore could be applied to EHR systems without the restriction of languages. BGLM is freely available at https://github.com/Bin-Chen-Lab/BGLM.

The Logical Observation Identifiers, Names and Codes (LOINC) is a common terminology used for standardizing laboratory terms. Within the consortium of the HiGHmed project, LOINC is one of the central terminologies used for health data sharing across all university sites. Therefore, linking the LOINC codes to the site-specific tests and measures is one crucial step to reach this goal. In this work we report our ongoing efforts in implementing LOINC to our laboratory information system and research infrastructure, as well as our challenges and the lessons learned. 407 local terms could be mapped to 376 LOINC codes of which 209 are already available to routine laboratory data. In our experience, mapping of local terms to LOINC is a widely manual and time consuming process for reasons of language and expert knowledge of local laboratory procedures.