3. Semantic interoperation of data where computer syste are
using different coding schemes. (See Table 17. 1.)
Types of Maps
Historically, data maps have been constructed employing a variety of approaches and can serve many different use cases. There is
no standard for classifying data maps, although the ISO technical
document does specify a set of quality indicators as noted in the
following section. A data map may be described minimally based
upon the protocols involved in creating map records, the nature of
the link used in the map, or by the characteristics of the terminology resources that are linked. Therefore, a descriptive map type may
be assigned as a list of one of each of the following descriptive sets.
Lexical mapping: Identifies sets of terms to be linked by
comparing words or character strings and is very dependent upon the language and dialect employed. The National Library of Medicine’s (NLM) MetaMap program
supports mapping of biomedical text to concepts in the
Unified Medical Language System (UMLS) and is an example of a lexical approach.
Semantic mapping: Employs the full defined meaning of
a concept in a source reference terminology or ontology to
identify equivalent or related concepts in the target. A reference terminology is a controlled terminology employing
a set of terms and relationships which capture and define
the meaning of each concept. The map of SNOMED CT to
ICD-10-CM discussed in the use cases later in the chapter
is an example of a map that was constructed semantically.
Nature of the Link Relationship:
Equivalence maps are designed to identify pairs of identical
concepts between the source and target. Semantic equivalence and lexical equivalence can lead to very different types
of map sets that may not serve the same use cases well.
Hierarchical (Taxonomic) maps order sets of concepts
Characteristics of the Source and Target:
as having more or less specificity when traversing from
source to target. They employ link relationships such as
“less-than,” “child-of,” “more-specific than,” or “is-a.” A
classification such as ICD-10-CM includes such implicit
relationships which specifies its hierarchical structure.
Knowledge-based (Rules-based) maps employ features of
coding context, business case, and workflow assumptions,
which apply to the interpretation and application of map
records, in order to create maps for terminologies where
the standards developer imposes constraints on the use
of the terminology scheme. The best example of a knowl-
edge-based map is the map of SNOMED CT to ICD-10-CM
in which World Health Organization (WHO) classification
guidelines and additional context from the patient demo-
graphics and problem list are employed by a set of rules to
map each SNOMED C T concept recorded as a problem (con-
dition) in the EHR.
Maps may be typed based upon the nature of the terminologies
involved in the map records, which may include classifications,
controlled terminologies, reference terminologies, or ontologies. A map such as the SNOMED CT to ICD-10-CM map may
be described as “reference terminology to classification.” ¢
1. Olsen, LeighAnne, Dara Aisner, and J. Michael McGinnis.
Institute of Medicine Roundtable on Evidence-Based Medicine. The Learning Healthcare System Workshop Summary.
Washington, DC: The National Academies Press, 2007.
2. Office of the National Coordinator for Health Information
Technology (ONC). “Connecting Healthcare for the Nation: A
Shared Nationwide Interoperability Roadmap.” 2017. https://
3. International Standards Organization (ISO). “ISO/TR 12300:2014
Health Informatics – Principles of Mapping between Terminolog-
ical Systems.” 2014. https://www.iso.org/standard/51344.html.
4. SNOMED International. “SNOMED CT Editorial Guide.”
James R. Campbell is a board-certified physician and a professor of
medicine at the University of Nebraska Medical Center.
Table 17. 1 Purposes of data map creation and interoperation of clinical terms
Table B Use Case Example
Data conversion Map EHR data from a legacy coding system to new standards Convert coded diagnosis records from ICD-9-CM to ICD-10-
Re-use Choose medication records from the EHR corresponding to
orders with a specific treatment reason
RXCLASS mapping of RxNorm medications to VA drug
classes to identify cancer therapies
Re-use Map SNOMED CT conditions to ICD-10-CM for reporting
morbidity and mortality
NLM release of SNOMED CT – ICD-10-CM rules-based map
Interoperability Extend and integrate terminology standards across domains
The SOLOR project