Controlled Terminology Courses
Public Review - Controlled Terminology
CDISC Controlled Terminology is maintained and distributed as part of the NCI Thesaurus on an NCI File Transfer Protocol (FTP) site and is available for direct download in Excel, text, odm.xml, pdf, html and OWL/RDF formats from the CDISC Controlled Terminology resources page on the National Cancer Institute website.
Controlled Terminology consists of question (e.g., Variables, TESTs and PARMs) and answer(e.g., response codelists, qualifier variable codelists), which are commonly referred to as codelists and are published alphabetically in the Controlled Terminology publication.
The terms within these codelists may have relationships to other terms within other codelists. For instance, a single TEST in the EGTEST codelist may have a finite set of responses located in the EGSTRESC codelist that constitutes a subset of the EGSTRESC codelist. Another instance, a single VSTEST value may have a constrained set of units of measure that are valid for the numeric responses to that VSTEST. These relationships are not readily apparent in the Controlled Terminology publication files.
To address this issue, the Controlled Terminology Teams have created Codetable Mapping Files based on published Terminology, which show relationships between terms in different_Controlled Terminology codelists. These supplemental files provide human and machine-readable linkages between published terms across multiple codelists and may be helpful for data QA/QC, CRF building, and data mapping. These files are for clinical use only.
The Controlled Terminology teams will continue to update these files as new Terminology is published, as well as develop new domain Codetable Mapping Files. If you are interested in seeing specific content developed, please submit the request through the New Term Request Site. CDISC is concurrently working on the development of electronically consumable formats of this content to be published out of CDISC Library.
Note: 2023-01-24: The SEND codetable mapping file has not been updated since CT Package 43. The SEND Leadership determined this file is not needed to support a SEND submission to regulatory authorities. The file will be removed effective May, 2023.
The Unified Code for Units of Measure (UCUM) contains a blueprint for the creation of compliant units of measure from more than 300 terminal unit symbols. UCUM is used in healthcare to populate electronic health records, such as laboratory records in LOINC, and in the ISO IDMP standard.
CDISC must specify a single, preferred unit of measure for the pharmaceutical industry to use in data submissions to regulators. Since UCUM does not control mathematical synonymy nor cover all units of measure required by the pharmaceutical industry, CDISC publishes submission values for units of measure across multiple codelists that support variables within CDISC findings, interventions, and trial design domains.
To seamlessly toggle between UCUM and CDISC Units, a mapping has been built and is updated quarterly with the CDISC Controlled Terminology publication. For each published CDISC submission value across all UNIT-based codelists in SDTM terminology, the mapping provides valid UCUM expressions associated with that unit concept. This allows a collected UCUM-compliant unit of measure to be easily matched to the appropriate CDISC submission value associated with that concept.
FDA requires the submission of Logical Observation Identifiers Names and Codes (LOINC®) within clinical LB domain datasets in new drug applications (NDAs), abbreviated new drug applications (ANDAs), and biologics license applications (BLAs) for studies starting after 15 March 2020 and for certain investigational new drugs (INDs) for studies starting after 15 March 2021. A recommendations document produced by a LOINC working group, composed of representatives from FDA, NIH, CDISC, and Regenstrief Institute, called for the development of a mapping file to map the most commonly submitted LOINCs in clinical research to the associated CDISC Controlled Terminology components associated with each LOINC part.
LOINC to LB Mapping File
This mapping file is intended to show examples of LOINC code mappings to CDISC variables and terminology to aid researchers’ adoption of the FDA requirement. The file is composed of LOINCs from the 2005 CDISC-LOINC mapping project, the LOINC Top 2000 SI, the LOINC Top 2000 US, as well as CDISC LB domain datasets submitted to FDA from 2014 - 2016. It includes more than 1400 LOINCs across the Chemistry, Hematology, Coagulation, Toxicology, Urinalysis, Serology, and other classes that would be mapped to the CDISC LB domain. The mapping file does not include LOINCs relevant to data that would be mapped to other CDISC domains such as microbiology, pathology, genomics, vital signs, fetal/neonatal screening, or administrative questions. For more information, please refer to the published ReadMe associated with the mapping.
CDISC Conformance Rules dictate that an <=8 character TESTCD/PARMCD and <=40 character TEST/PARM value be available for data submission to regulators. Two distinct codelists are created to support two distinct, paired submission values, however the underlying semantic meaning remains the same. It is not easy to tell which paired submission value goes with which, as the paired submission values exist on different rows of the CDISC Terminology publications, sometimes quite far apart. The CDISC Synonym column is not helpful to ascertain the paired 'decode' value when there are multiple synonyms. Currently, NCI C-codes can be used to pair TEST/TESTCD or PARM/PARMCD submission values but this requires manual manipulation of the spreadsheets. Currently users must manipulate the existing formats of CDISC Controlled Terminology produced by NCI-EVS to generate a more integrated view of these paired values.
NCI-EVS has produced a view of these paired codelists that show the paired submission values (TEST/TESTCD or PARM/PARMCD) for each terminology subset (SDTM, SEND, and ADaM) on the same row of data. Only SDTM, SEND, and ADaM files have been produced because these are the only CDISC Terminology subsets that contain paired codelists. These files contain only paired codelist values; they do not contain all terminology values for SDTM, SEND, and ADaM. NCI-EVS will update these files, as appropriate, with the CDISC quarterly terminology release cycle.