Evaluating the Coverage of LOINC Hematology Data Dictionary Wanchana Pontongmak MSc. Thai Health Information Standards Development Center (THIS) Nonthaburi, pontongmak@gmail.com Boonchai Kijsanayotin M.D., PhD. Thai Health Information Standards Development Center (THIS) Nonthaburi, boonchai@hsri.or.th Abstract Background: At present, share-care is needed in every health organization. To achieve to the share-care, health data standards are needed. Thailand needs standard for identifying laboratory tests to enable the complete healthcare system. The Logical Observation Identifiers names and codes (LOINC) was suggested by eHealth experts in country as the most suitable standard for laboratory. The LOINC has potential to enable the interoperability in Health Information System (HIS) in country. LOINC is a universal terminology standard for identifying laboratory and clinical observation. Regenstief institute mentioned that LOINC is the most comprehensive standard for laboratory which covers more than 99% of laboratory tests in United States. The latest version of LOINC contains more than seventy thousand terms. Standardization by mapping local laboratory tests to LOINC is very laborious and tedious due to many reasons such as knowledge domain, complexity of standard, especially, the large number of LOINC. The Hematology Data Dictionary (HDD) has potential to minimize these problems in domain of hematology laboratory. However, the coverage of the hematology data dictionary is still unclarified. Evaluating the coverage of hematology data dictionary will ensure the reliable and increasing user confidence when using this tool. Objective: To evaluate the coverage of the hematology data dictionary by using hematology test lists from three hospitals in Thailand. Materials and Methods: The hematology data dictionary was tested the coverage by mapping local hematology tests from three hospitals to the LOINC codes in the hematology data dictionary. The mapping number and percentage were calculated and the precision and recall rates of the hematology data dictionary were measured as well. Results: 88.06%-94.23% of hematology test lists from three hospitals could be assigned by LOINC codes in the hematology data dictionary. The precision rate was 100% and the recall rate was 92.72% Discussions and Conclusions: The high percentage of mapped codes represents that LOINC is appropriate for Thai laboratory. The high recall rate indicated that the hematology data dictionary is suitable to use as a tool for standardizing local laboratory tests in hematology laboratory of Thailand. Keywords- LOINC; Standard; Data Dictionary Interoperability; Hematology Introduction Logical Observation Identifiers Names and Codes (LOINC) is a terminology standard that has been developed by The Regenstrief Institute, Indianapolis in United States of America. It is a universal names and codes system for encoding clinical observation and medical laboratory. LOINC, which released on December 2013 contains 72,625 terms, is broadly used in 162 countries and free of charge[1]. The Regenstrief Institute also provides mapping tool, the Regenstrief LOINC Mapping Assistant (RELMA) that can be used for browsing LOINC terms in LOINC database as well as for mapping local terms to LOINC[2]. The data dictionary with standard laboratory terminology has potential to standardize laboratory data. The laboratory data dictionary is a minimum standard test list of laboratory tests. The need of data dictionary greatly increases when the share-care are popularizing[3]. To achieve the interoperability in Laboratory Information System (LIS), the data standards of laboratory are needed. Currently, Thai healthcare systems interested to apply LOINC terminology for laboratory test data[4]. However, the standardization of data in the field of medical laboratory is a laborious work. Hematology is the science of blood, blood-forming organs and blood diseases. The hematology discipline is concerned with the blood disorders and malignancies treatment such as hemophilia, leukemia, and anemia. Hematology tests are often required at the first place by clinicians in order to diagnose patient illness[5, 6]. the usefulness of hematology information is not only for clinical diagnosis but also for treatment and others area e.g., epidemiology report, resources and human management, reimbursement, and clinical research. Standardizing hematology data by mapping local hematology test with laboratory terminology such as LOINC will extend the usability of hematology data especially the patient care across different health organizations. Unfortunately, the standardization task is always suffering and tedious[7, 8]. The hematology data dictionary is a minimum standard code set of hematology test which was created by using hematology tests from two large university hospitals in Thailand mapped to LOINC terminology. The hematology data dictionary could be facilitated healthcare facilities to standardize local hematology tests into LOINC terms correctly and easily[9]. To clarify the usability of the hematology data dictionary, the evaluating coverage of hematology data