2.1 General Naming Conventions

2.1.1 Abbreviations in names of Component (Analyte)

In general, abbreviations are not used in the Component. For example, we use “total”, “fraction”, “oxygen”, “Alpha”, and “Beta” rather than “tot”, “frac”, “O2”, “A-“, and “B-“. However, there are a few exceptions, such as the commonly used abbreviations shown in Table 2.

Table 2: Example Component Abbreviations

2.1.2 General naming rules for the Component (Analyte)

1. Place the identifier of the substance being measured first. This means “Hepatitis A antibodies (Ab)” not “Antibodies, Hepatitis A.”
2. Use the generic name of a drug, not the brand name, when referring to drug concentrations and antimicrobial susceptibilities, e.g., Propranolol, not Inderal. We will usually include the brand or trade names in the related names (synonyms) field.
3. Use full taxonomic name of an organism or virus name (not the disease) when describing a test that diagnoses that disease. Say “Rickettsia rickettsii Ab” not “Rocky Mountain spotted fever Ab”. Say “herpes simplex virus Ab” not “HSV Ab.” The disease name should be included as a synonym in the related name field.
4. Species and groups of species: SP identifies a single species whose identity is not known. SPP identifies the set of species beneath a genus. We have a third case, however. In some tests, antibodies apply to different strains of species. In rickettsial diseases, the antibodies are then against groups of species, e.g., the spotted fever group or the typhus group. In this case we use Rickettsia spotted fever group and Rickettsia typhus group.
5. When tests include the name of a bacterium (e.g., Neisseria gonorrhoeae DNA probe) for the formal LOINC name we use the full bacterial name from the International Journal of Systematic and Evolutionary Microbiology. When it includes the name of a virus (e.g., West Nile Virus IgM antibodies), we use the viral name as given by Index Virum.
6. When the test measures an antigen to a specific species of organism but cross-reactivity is such that other organisms are identified, the name should be the principal organism that is targeted by the test.
7. Avoid “direct” and “indirect” except as parts of synonym names. Avoid “conjugated” and “unconjugated” when a more precise term, such as “glucuronidated” or “albumin-bound” is available.
8. Use “platelets”, not “thrombocytes.”
9. Name vitamins by the chemical name. For example, use thiamine not Vitamin B1, The name containing “Vitamin” will be included as a synonym. This is the only reasonable approach because all vitamins have a chemical name but not all vitamins have a “numbered” vitamin name. One exception to this rule is that we will use the widely accepted vitamin name for nutrition intake terms.
10. Always specify whether serology tests measure the antigen or antibody, using the abbreviation “Ab” for antibody and “Ag” for antigen. Remove the “anti” from “ANTI X Ab.” It is redundant and obscures the most significant word in the name. Thus, “anti-smooth muscle Ab” becomes “Smooth muscle Ab.” Common abbreviations or shortened names, e.g., ANA for anti-nuclear antibody, will be found in the related names field.
11. VDRL will be named Reagin Ab because that is what it is. We will have to depend upon synonyms and aliases to equate our “standardized” names with the old names.
12. Use the noun form of the target of the antibody, e.g., Myocardium Ab, not Myocardial Ab.
13. Anion vs. acid: Always use the anionic name for chemicals, not the acid name, e.g., lactate, citrate, and urate, not lactic acid, citric acid, and uric acid. The acid form of the name will be included in the related names field of the database.
14. Alcohols: Always use the single-word names for alcohols: methanol, not methyl alcohol; ethanol, not ethyl alcohol, and so on.
15. Always spell out OH as Hydroxy, or as – ol, with no space or hyphen between Hydroxy and the next word.
16. Greek letters, alpha, beta, gamma, etc., are always spelled out (e.g., alpha tocopherol, not A-tocopherol), with a space between the spelled out Greek letter and the rest of the chemical name.
17. Use pH, not log (H+).
18. Whenever possible, the Component will contain the scientific names of allergens. Note: This is a convention implemented in January 2002.
19. Avoid use of the word “total” in laboratory test names, except when denoting the denominator of a fraction. Thus it is Alkaline phosphatase, not Alkaline phosphatase.total, but Alkaline phosphatase.bone/Alkaline phosphatase.total.
20. For drug metabolites, we will generally use the “nor” form rather than “desmethyl”, e.g., nordoxepin not desmethyldoxepin. When the distinction between specific forms of the compound is important, however, we will make an exception, e.g., O-desmethylvenlafaxine and O-desmethylencainide.
21. Abbreviate units of time, such as Y for year, D for day, and W for week. This only applies to time as a unit of measure, e.g., in the past 7D, and not when year or day are used to refer to a date or general time construct, e.g., next year.

2.1.3 Punctuation in Analyte names

A number of Analyte names include punctuation characters such as commas, for example, to identify the position of multiple alkyl groups in a carbon chain. We will avoid special characters, e.g., commas, dashes, and parentheses, except where they are included in the name specified by IUPAC, the Chemical Abstract Service (CAS) convention, or another international convention. So for example, commas will appear in multiple substitutions of alkyl chains per the CAS standard, dashes, asterisks and colons will appear in HLA allele names, colons will appear in the names of some microorganisms, and parentheses (i.e., round brackets) will appear in the names of red blood cell antigens.

2.1.4 Case insensitivity

All names are case insensitive. Prior to December 2006, we used upper case in the database and our examples, but we then changed to mixed case for easier readability. In electronic messages senders and receivers can use upper, lower or mixed case. However, the meanings should not be sensitive to case conversions to avoid any possibility of confusion when the information is sent over networks that may apply case conversion. To identify parts of the few names that by international convention are case sensitive, such as red blood cell antigens, we use the word “little” in front of the letter that is lower case. We use a similar convention to indicate superscripts with the word SUPER. See examples in Table 3.

Beginning in June 2019 with LOINC version 2.66, we adopted tall man lettering for drugs with similar names based on the list recommended by the Institute for Safe Medication Practices, which includes the list previously published by the U.S. Food and Drug Administration. Tall man lettering uses upper case letters to distinguish key parts of the drug name in order to help visually differentiate similar names, such as cycloSPORINE and cycloSERINE. This format is still case insensitive in that the meaning is the same regardless of the case; however, for senders and receivers that can support mixed case display, the tall man lettering helps draw attention to the discriminating parts of similar names.

Table 3: Example Case Specifying Conventions

2.1.5 Roman numerals vs. Arabic numerals

Whenever possible, numerals shall be represented in their Arabic form. However, when the conventional name uses Roman numerals, as is the case for clotting factors such as factor VIII, the LOINC primary name will use Roman numerals and we define a synonym containing Arabic numerals.

2.1.6 Exponents

We will use the phrase “exp” to indicate an exponent in the Component. According to the Unified Code for Units of Measure (UCUM), exponents are typically represented using an asterisk (*) or carat (^). However, asterisks and carats both already have defined meanings in LOINC. (As described later, asterisks are used in variables that report another specific part of the LOINC name, and carats are used as delimiters of subparts.) For this reason, we use “exp” to avoid ambiguity. For example, the concept “height raised to the power of 2.7” is represented as “height exp 2.7”.

2.1.7 Use of the slash (/)

In the Component, a slash (/) is used to distinguish between a numerator and a denominator (a.k.a. Divisor) for terms that represent fractions or ratios. For example, “Basophils/100 leukocytes” represents the fraction of basophils out of the parent population of 100 leukocytes, and “Lecithin/Sphingomyelin” represents the ratio of lecithin to sphingomyelin. In the System, a slash is used as a conjunction and means “or”. A common example is “Ser/Plas”, which means that either Serum or Plasma is a suitable specimen for measuring a particular analyte.

2.1.8 Use of curly braces {} in LOINC names

In Section 2.5 we describe the use of XXX in the System when the material is unknown or specified elsewhere in the HL7/ASTM message. In some domains, particularly clinical observations, we have adopted a newer style of notation with curly braces to indicate that the information is provided elsewhere.

For example, LOINC term 32491-3 represents a deep tendon reflex observation where the anatomic location is not specified in the term name:

Deep tendon reflex:Find:Pt:{Reflex location}:Ord:Observed

Such terms could be used in post-coordinated expressions where the value of “reflex location” is communicated elsewhere. The curly braces notation has the advantage over using XXX in some cases because it allows a more precise indication of the set of possible entities to expect for that piece of information. In the reflex example above, saying {reflex location} narrows the expected set of possible values down to those anatomic regions where a muscle tendon reflex can be observed.

We have made use of this notation primarily in the System and Method part of the name.

A standing example of this convention applies to microbiology molecular detection terms (Class: MICRO) matching Property = PrThr, Time = Pt, Scale = Ord: these terms use {Specimen} for System and {Molgen} for Method, regardless of which specimen type or molecular technique is actually used. For example, a DNA/RNA detection term for a given pathogen in this pattern would read:

[Pathogen] DNA:PrThr:Pt:{Specimen}:Ord:{Molgen}

2.1.9 Resolving post-coordinated values via defining LOINCs

Some curly-brace axis values are resolved to a narrower, specific value at the point of use rather than through a distinct pre-coordinated LOINC term for every combination. This avoids unbounded combinatorial growth of terms as specificity needs grow (e.g., gene × detection- technique combinations for {Molgen} above).

The mechanism uses only files and columns that already exist in the LOINC release — no new tables or schema changes:

1. The curly-brace axis value (e.g., {Molgen}) flags the value as post-coordinated.
2. The Part behind that axis value carries a row in the PartRelatedCodeMapping file with ExtCodeSystem = http://loinc.org (rather than an external system like SNOMED CT), pointing at a defining LOINC — the term that governs which values are valid substitutes.
3. The defining LOINC carries an Answer List (via the LoincAnswerListLink file).
4. Each valid narrower resolution is listed as a row in the AnswerList file, with ExtCodeSystem = http://loinc.org and ExtCodeId pointing at the Part for that resolution.

A defining LOINC uses the FSN format Value:-:-:-:Nom: and is classified under the LOINC.DEFINING class, distinguishing it as a modifier for primary observations rather than an ordinary observable term. As of this release, only prototype defining LOINCs for Specimen and Molgen have been created; broader rollout is pending further modeling guidance.

See Section 14 for a full description of post-coordination in LOINC, including the entity-relationship model behind this mechanism and guidance for implementers.