Logical Observation Identifier Names and Codes (LOINC(tm)) Users Guide vs. 1.0 Introduction LOININTR.TXT Revised 4/28/96 ---------------------------------------------------------- Please send questions and comments to: Regenstrief Institute c/o Kathy Hutchins 1001 W. 10th St., RG-5 Indianapolis, IN 46202 Internet: loinc@regenstrief.iupui.edu ---------------------------------------------------------- This and other relevant documents are available via FTP/Gopher: dumccss.mc.duke.edu/standards/HL7/termcode/loinclab WWW URL: http://dumccss.mc.duke.edu/ftp/standards/HL7/termcode/loinclab/ The file names are: Description File Name ----------- --------- LOINC data base (WP 6.0) LOINDBW1.WP6 uncompressed LOINC data base (WP 6.0) LOINDBW1.ZIP PKZIPped LOINC data base (ASCII) LOINDBT1.TXT uncompressed LOINC data base (ASCII) LOINDBT1.ZIP PKZIPped LOINC data base (WinWord 2.0) LOINDBW1.DOC uncompressed LOINC data base (WinWord 2.0) LOINDBWW.ZIP PKZIPped LOINC data base (DBF) LOINCDBF.EXE Self-extracting file LOINC Users Manual (WP 6.0) LOINMAN1.WP6 LOINC Users Manual (WinWord 2.0) LOINMAN1.DOC LOINC Introduction (ASCII) LOININTR.TXT (this file) LOINC Read Me (ASCII) LOINRDME.TXT ---------------------------------------------------- Copyright and Terms of Use -------------------------- Copyright 1995 Regenstrief Institute and the Logical Observation Identifier Names and Codes (LOINC(tm)) Committee. All rights reserved. Permission is hereby granted, without written agreement and without license or royalty fees, to use, copy, or distribute the LOINC codes, LOINC Users' Guide, and the contents of the LOINC database for any purpose, so long as this copyright notice appears on any copies of the LOINC database and Users' Guide, and that the following conditions are met. Users of the LOINC codes agree to the following conditions: 1) They will not change the meanings of any of the LOINC codes. 2) They will not change any contents in the defined LOINC Fields. (Users may add their own new fields to the data base if they want to attach additional information to the existing LOINC record.) 3) They will include this notice of copyright and terms of use in any copies of the LOINC database that they distribute. 4) If new records are added to the LOINC database as distributed to deal with local requirements, these records must be assigned a LOINC code number containing a leading alphabetic "X" so that the new term cannot be confused with new LOINC codes as they are assigned by the LOINC committee. 5) Those who incorporate any part of the LOINC data base into another laboratory test definition data base for distribution outside of their corporation must include the LOINC number (field #1) all six name fields (#2-7), the related terms (field #8), and the answer list (field #18), and include this copyright notice on the electronic document that incorporates the LOINC database. Regenstrief Institute and the members of the LOINC Consortium do not accept liability for any omissions or errors in the LOINC data base, and all EXPRESS AND IMPLIED WARRANTIES, INCLUDING THOSE RELATING TO MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE DISCLAIMED. LOINC is a trademark of the Regenstrief Institute. This legend should be displayed on every copy of the database (both on the diskette itself and in a text file loaded onto the diskette or onto the Internet), and on all user manuals and other materials used in connection with the LOINC data base. Preface and Introduction ------------------------ The LOINC database provides a set of universal names and ID codes for identifying laboratory test results. The purpose is to facilitate the exchange and pooling of clinical laboratory results, such as blood hemoglobin or serum potassium, for clinical care, outcomes management, and research. Currently, many laboratories are using ASTM 1238-94 or its sister standard, HL7, to send laboratory results electronically from producer laboratories to clinical care systems in hospitals. Most laboratories identify tests in these messages by means of their internal (and idiosyncratic) code values, so the receiving systems cannot fully "understand" the results they receive unless they either adopt the producer's laboratory codes (which is impossible if they receive results from multiple source laboratories, e.g.; the hospital lab, the local commercial lab, and a nursing home lab), or invest in work to map each laboratory's code system to the receiver's internal code system. If laboratories all used the LOINC codes to identify their results in data transmissions, this problem would disappear. The receiving system with LOINC codes in its master vocabulary file would be able to understand and properly file HL7 results messages that also use the LOINC code. Similarly, government agencies would be able to pool results (within limits) for tests from many sites if they were reported electronically using the LOINC codes. The LOINC codes (and names) for test observations should be of interest to hospitals, clinical laboratories, doctors' offices, state health departments, governmental health care providers, third party payors, and organizations responsible for quality assurance and utilization review. The LOINC codes are not intended to transmit all possible information about a test. They are only intended to identify the test result. Other fields in the message will transmit the identity of the source laboratory and the very detailed specimen information. (For instance, the code may identify a blood culture, but the message source code can be more specific and identify the specimen as pump blood.) The level of detail in the LOINC definitions was intended to distinguish tests that are usually distinguished as separate test results within the master file of existing laboratory systems. Indeed, we are using the master files from seven different U.S. laboratories to shape this effort. Each LOINC record corresponds to a single test result. (The LOINC database does not yet deal with orderable sets of tests such as CHEM 12.) The record includes fields for specifying: 1) The analytes (e.g., potassium, hemoglobin, hepatitis C antigen). 2) The property of the measurement (e.g., is it a mass, a mass concentration, an enzyme (catalytic rate). 3) Whether the measure is a point observation or an observation integrated over time (e.g., 24 hour urine). 4) The type of specimen (e.g., urine, blood, serum). 5) The general precision, e.g., whether the measure is quantitative (a number) or qualitative (red, blue, green). 6) Where relevant, the method used to produce the test. It also contains information about the amount, route, and timing of challenges. (e.g., Glucose^1H post 100 dl glucose p.o.) The LOINC identifiers do not usually include the method in the name for chemistry tests, but do include it for most serologic tests and coagulation studies because methods tend to be less important in chemistry where many tests are standardized to normalized methods. This same distinction is usually reflected in the master files of operating laboratories. Of course, the method can be reported as a separate item of information in a result message regardless of whether it is part of the test name. We have used many sources for constructing the database, including the Silver Book from International Federation of Clinical Chemistry, clinical pathology (e.g., Henry and Tietz), the expertise and work of the LOINC members, and Euclides. We have also reviewed the master test files of seven sources (Indiana University/Regenstrief, University of Utah ARUP, Mayo Medical Laboratories, LDS Hospital in Salt Lake City, the Department of Veterans Affairs, MetPath-Corning Laboratories, and University of Washington). This has been an empirical effort. Our goal is provide codes to which laboratories could map to their master files. The database includes fields for each of the six components of the name, and in many cases to also include Euclides codes (for the analyte), International Union of Pure and Applied Chemistry/ International Federation of Clinical Chemistry (IUPAC/IFCC) codes, and ASTM codes, as well as related words, synonyms, and comments. We plan to map the LOINC code to SNOMED and other code synonyms as well. Related words ("synonyms") are included to facilitate searches for particular laboratory test results. Laboratories and medical record system users should record the LOINC codes as attributes of their existing master files and use the LOINC codes and names in the OBSERVATION ID field of the ASTM and HL7 OBX segment to identify laboratory results. The LOINC data base is presented to you as an electronic document grouped by "sensible" categories to make it easier to find general areas of interest. We have the usual categories of chemistry, hematology, serology, microbiology, and toxicology; we also have categories for drug levels and blood counts (the observations you would find reported on a CBC and differential, counts of other fluids such as cerebrospinal fluid, peritoneal fluid, etc). The urinalysis category contains the both observations reported on an urinalysis and reports about crystals from other specimens here. We have separated antibiotic susceptibilities into their own category. We have a small category for virology studies that include methods for identifying viruses (isolated and antigen), and also one called clinical where we collect observations that come from the requestor but are reported back by the laboratory (e.g., inspired O2). Table 1 lists all of these classes in detail. There is nothing sacred about these categories. You will be able to sort the database by whatever class is convenient when you get the electronic version. TABLE 1: Classes ---------------- ABXBACT Antibiotic susceptibility BC Cell counts (blood, CSF, pleuretic fluid) BLDBK Blood bank CHAL Challenge tests CHALSKIN Skin challenge tests CHEM Chemistry CLIN Clinical terms for reporting patient clinical state COAG Coagulation study DRUG Drug levels DRUGDOSE Drug dose (for transmitting doses for pharmacokinetics) FERT Fertility HEM Hematology (excluding coagulation & differential count) HLA HLA tissue typing antigens MICRO Microbiology (Former BACT, PROBE, VIRO, and infectious agent SERO) SERO Serology (antibodies and antigens except blood bank and infectious agent) TCELL T Cell surface models TOX Toxicology UA Urinalysis We are aware of gaps and areas that are not complete. For example, we have not yet addressed the subject of surgical pathology or cytology. (These should be covered within the next four to six months.) We have only included a "not specified" observation identifier with "unspecified" specimens for common laboratory chemistry tests; more may be needed. A few hematology records have no type of property. We have only addressed results, not orders. In only a few cases have we suggested standard answer lists for tests that are recorded as codes. We have defined fields in the data base for a number of data elements, e.g., example units, sample normal ranges, but none of those fields are filled in. We hope to greatly enrich the existing synonym and answers list in the next version. For some kind of tests, the database provides two ways to report the values. For example, blood cell antigens might be presented as a "panel" with separate "tests" which report each possible antigen as present or absent responses if the test is to establish paternity, for crossmatching, the result would be reported as a list of antigens found. We try to provide for both methods of reporting in the LOINC database by including codes for both types of test identifiers. The Regenstrief Institute and the LOINC committee will maintain the database while grant support is available (at least until October 1997). We expect to find a long term home for the database before then. The LOINC database (which identifies over 6,000 different lab tests) and supporting documentation will be available through the Duke HL7 standards data list server. The LOINC database is stored in seven different file formats. In each of them the first part of the file contains the copyright notice with permission to use the database for any purpose without charge or written permission. We have copyrighted the database and this document to assure that multiple variants of the codes do not emerge. Having many variants would defeat the purpose of a universal test result identifier. Tab Delimited ASCII: Each record of the database is on a separate line. Each record is terminated by CR/LF, and each field is delimited by a tab character. Fields are enclosed in double quotes ("). Spreadsheets and database programs can easily handle such files, but it would result in an awkward word proecessing document. The ASCII file is the "database of record" -- unlike the word processing versions, it will always contain all implemented fields. WordPerfect 6.0 file: This file is formatted to print landscape in a Courier 6 point font. Some of the longer fields float vertically. This will result in an easily read document, but it is basically useless for input into databases or spreadsheets. The dimensions of the printed page make it impossible to display all database fields in this file. WinWord 2.0 file: This file was produced by writing out the WordPerfect file through the WinWord export filter. It has been loaded into WinWord and examined; the translation is usable but not perfect. If your system does not contain the fonts used in the original file, the columns may be misaligned. Consult the Word for Windows Help topic "Improving compatibility with a document created in a different file format" for hints on dealing with this. Import of this file into Word for DOS has not been tested. All three of these files are available both zipped and unzipped. (PKUNZIP v. 2.04 or compatible required) DBase format: A DBF file and accompanying index (.CDX) file which enables sorting on each of the 32 fields in the data base is distributed as a self-extracting ZIP file, LOINCDBF.EXE. A documentation file is also included. The LOINC Users Guide is available both as a WordPerfect 6.0 file or a WinWord 2.0 file. This Users Guide explains the structure of the database, its rationale, and the rules we used for naming test results. It is not compressed. The Winword file, like the WinWord database file described above, was created with the WordPerfect export filter, and like the database file, it is usable but not perfect. In particular, table of contents tags are replaced with "ERROR: BOOKMARK NOT FOUND" and the automatically generated section numbers are incorrect. Tables are sometimes misaligned but are still readable. See the preceeding paragraph for font translation information. Import of this file into Word for DOS has not been tested. The introduction to the Users Guide (this document) is available as a separate ASCII text file. We welcome corrections and/or extensions to the database. We are not interested in adding terms that might be needed in some future situation but we are very interested in adding test observations that are actively being reported today. If you wish to submit terms that are not yet included, please put terms in the LOINC format as described in this guide and exemplified in the LOINC data base. Please include a letter describing your role in laboratory computerization, the planned use for these terms, and your mailing and email address. (If you submit more than 10 terms, please submit them as ASCII documents, one term per line and tab delimited, or as a DBASE file). Fewer than ten terms may be submitted on hard copy. If you have a big submission in mind, it would be best to contact one of the LOINC committee members to discuss the project and to submit a trial set before you embark on the full project. In all cases of the test name-field, you must complete at least the first (name), third (duration), fourth (specimen), and fifth (level of precision) components of the name. We will help with the second part of the name (kind of property). You must also include synonyms and a description of the variable. Become completely familiar with the LOINC database, so you can be sure the term(s) of interest is not already included, and use the LOINC Users Guide to build the entry for each of the specified fields. We are considering expanding the scope of this effort to other clinical variables (such as hospital measurements, including: vital signs, CNS pressure monitoring, intake and output, EKG measurements, obstetrical ultrasound measurements, etc.). We are looking for volunteers with expertise on these subjects to help us. If we can find the right volunteers we will pursue such extensions. When writing, provide a description of your background and interest. Mail correspondence to: Regenstrief Institute for Healthcare c/o Kathy Hutchins 1001 W. 10th St., Indianapolis, IN, 46202 Internet: loinc@regenstrief.iupui.edu. LOINC COMMITTEE MEMBERS ----------------------- Clement J. McDonald, M.D., Chairman Regenstrief Institute and Indiana University School of Medicine, Indianapolis IN John Baenziger, M.D. Indiana University Hospital, Indianapolis, IN Linda Charles, Ph.D. Quintiles, Morrisville, NC Georges DeMoor, M.D. University Hospital Ghent, Ghent, Belgium Diane Dwyer Maryland Department of Health and Mental Hygiene, Baltimore, MD Tom Fiers, M.D. University Hospital Ghent, Ghent, Belgium Arden Forrey, Ph.D. University of Washington, Seattle, WA Brian Griffin MetPath Labs, Westwood, MA Stan Huff, M.D. IHC/Primary Children's Medical Center, Salt Lake City, UT Dennis Leavelle, M.D. Mayo Medical Laboratories, Rochester, MN Diane Leland, Ph.D. Riley Hospital for Children, Indianapolis, IN Doug Martin, M.D. Roudebush VA Medical Center, Indianapolis, IN Frank Stalling, M.D. Dallas ISC Dept.of Veterans Affairs, Grand Prairie, TX Anders Thurin University Hospital, Linkoping, Sweden Wayne Tracy SpaceLabs Medical Inc., Overland Park, KS Ann Tullis Indiana University Hospital, Indianapolis IN Larry Widman University of Oklahoma Health Sciences Center, Oklahoma City, OK Acknowledgements ---------------- We wish to thank Henrik Olesen, Chairman of the IUPAC Commission on Quantities and Units in Clinical Chemistry, for his very helpful comments and insights about laboratory test coding. This endeavor was supported in part by grants and contracts from the John A. Hartford Foundation of New York, the National Library of Medicine (Contract NO1-LM-3-3410), and the Agency for Health Care Policy and Research (AHCPR) Grants HS 05626 and HS 07719-013). Much of the work was performed in the Regenstrief Institute.