Regenstrief Institute is pleased to announce the availability of the Common Lab Orders LOINC Value Set Version 1. This value set was developed by the Lister Hill Center at the National Library of Medicine, with the help of the Regenstrief Institute and LOINC Committee. The list is sorted by LOINC class and within class alphabetically by LOINC long common name. This list is referenced by the Healthcare Information Technology Standards Panel (HITSP) C80 Clinical Document and Messaging Terminology Construct in Table 2-97 “Laboratory Order Value Set” which states that it “should be considered a minimum “starter” set,” and “does not attempt to include all possible LOINC codes” or all possible lab orders.

This list has evolved over the last 6 months. The original list was derived from frequency distributions of two large, and two smaller, real life databases and that had mapped all of their LOINC order and/or result codes to LOINC. These databases included a frequency distribution of 10 million test orders from the Indiana IHE, a similar analysis of 30 million test results from United Health Care; 1.5 million test orders from five NE US hospitals, 200,000 tests orders from a SE US hospital and a list of 200 tests gathered by survey from an internal medicine project run by Stasia Kahn, MD (Chicago). The first four of these data sets had LOINC codes attached; we added LOINC codes for the tests from Dr. Kahn’s list. The Indiana Health Information Exchange (HIE), NE US and SE US lists were weighted toward hospital laboratories (including inpatient and outpatient). The United Health Care results come from a national database and include only outpatient results produced at commercial laboratories – these help balance to the hospital orientation of the other samples. We started with the Indiana source and only added terms to the “99-percentile” list if they were not already included. The original set comprised about 300 tests and covered close to 99% of each of the sources’ laboratory volumes.

We presented that original table of common tests at the HITSP face to face meeting in Silver Spring on November 5th, after which it was sent out for public comment. We received a number of useful comments about the original database. To resolve these comments, we have had a number of very productive and useful meetings with the American Clinical Laboratory Association regarding the usage and definitions of some of these tests within the industry. As a result of those meetings and their expertise, we have replaced some items that were mis-conceptualized, reviewed and clarified the definitions of many panels, and added a set of tests based on the statistics of ACLA labs. We have removed a handful of tests that were incorrectly included as orders. We have not yet finished the detailed review with ACLA, and so have also held back about
60 tests (including some very common ones) from the original list. We will be meeting with ACLA to work through and polish these remaining tests. We expect to update this list accordingly within the next month or two, and updated versions will be available at http://loinc.org/usage

We’d like to thank to all who contributed data, helped develop this list and provided such valuable feedback. Special thanks to the HITSP Care Management and Health Records Domain Technical Committee, and the ACLA members who have provided extensive analysis.

Your comments and suggestions are welcome: commontestsandorders@mail.nih.gov