The ICD-10-CM code set is governed by a compact but demanding set of conventions — rules baked into the Tabular List and the Alphabetic Index that tell you how characters, symbols, and instructional notes must be read. These conventions are codified in Section I.A of the ICD-10-CM Official Guidelines for Coding and Reporting, maintained jointly by CMS and the National Center for Health Statistics (NCHS). Miss one — a stray placeholder, a misread "with," an Excludes1 that should have blocked a pairing — and an otherwise reasonable code becomes a denial or an audit finding. This guide walks through the conventions coders trip over most often: the seventh character, the placeholder X, the "code first" and "use additional code" sequencing rules, and the NEC/NOS abbreviations. For the underlying anatomy of a code, pair this with our companion piece on ICD-10-CM code structure.
The Seventh Character
Certain ICD-10-CM categories require a seventh character to complete the code. It is not optional detail — where the Tabular List calls for it, a code without the seventh character is invalid and will not be accepted on a claim. The seventh character appears most heavily in Chapter 19 (Injury, poisoning, and certain other consequences of external causes, S00–T88), in Chapter 15 (Pregnancy, childbirth and the puerperium), in some musculoskeletal categories in Chapter 13, and in the glaucoma codes of Chapter 7.
For most injury and external-cause categories, three values dominate:
- A — initial encounter. The patient is receiving active treatment for the condition: surgery, an emergency-department visit, evaluation and continued treatment by a new physician. "Initial" refers to the phase of care, not the number of visits.
- D — subsequent encounter. The patient has finished active treatment and is now in the healing or recovery phase — cast changes, medication adjustment, routine aftercare.
- S — sequela. A residual condition that arises as a direct consequence of the original injury, such as a scar contracture following a burn. Sequela coding usually needs two codes: the condition produced (the sequela) sequenced with the injury code carrying the seventh character
S.
Fracture categories carry a richer alphabet. The seventh character on a code such as S52.501 (unspecified fracture of the lower end of the right radius) encodes both the episode and the fracture's status: A for initial encounter for a closed fracture, B for initial encounter for an open fracture, D for subsequent encounter with routine healing, G for subsequent encounter with delayed healing, K for subsequent encounter with nonunion, P for subsequent encounter with malunion, and S for sequela. So S52.501A is a valid, billable seven-character code, while S52.501 on its own is incomplete.
Chapter 15 uses the seventh character differently: for many pregnancy codes it identifies which fetus a condition applies to in a multiple gestation — 0 for "not applicable or unspecified," and 1 through 9 to single out a specific fetus.
The Placeholder X
The seventh character must always occupy the seventh position of the code — never the fourth, fifth, or sixth just because those slots happen to be empty. When a category requires a seventh character but the code itself is shorter than six characters, ICD-10-CM inserts a placeholder X to fill the intervening positions and push the seventh character into place.
The classic example is a poisoning code. T36.0X1A means "Poisoning by penicillins, accidental (unintentional), initial encounter." Read it position by position: T36.0 is the penicillins substance, the X in the fifth position is a pure placeholder with no clinical meaning, the 1 in the sixth position captures the intent (accidental), and the A in the seventh position is the episode of care. Drop the X and the code collapses to an invalid five- or six-character string.
The placeholder also appears mid-code when a category simply has no defined character for a given position but still requires the seventh. Many Chapter 15 codes look like O33.3XX0 — "Maternal care for disproportion due to outlet contraction of pelvis," with two placeholder Xs holding the seventh-character fetus identifier in position. The rule to internalize: the X is structural, not descriptive, and it is never omitted. A quick way to confirm you have the placeholders and seventh character right is to look the code up in an ICD-10 code browser, which shows the complete valid form and flags whether the code is billable as typed.
NEC and NOS
Two abbreviations recur throughout the classification and are frequently confused, because both signal a lack of a perfectly matching code — but for opposite reasons.
NEC — Not Elsewhere Classifiable
NEC means the provider documented a specific condition, but the classification has no distinct code for it. It is the "other specified" situation: the diagnosis is known and detailed, yet ICD-10-CM lumps it into a residual "other" code because no more precise entry exists. In the Alphabetic Index, NEC directs you to a code that typically ends in .8 ("other specified"). Reaching for an NEC code is a statement that you have specific documentation and the code set simply has not carved out a slot for it.
NOS — Not Otherwise Specified
NOS is the equivalent of "unspecified." It signals that the documentation lacks the detail needed to assign a more precise code — the intent, laterality, severity, or type is simply not recorded. NOS codes commonly end in .9 ("unspecified"). The distinction matters clinically and financially: an NOS/unspecified code can flag incomplete documentation and, in risk-adjusted or quality-reported settings, may pay less or fail a quality measure. NEC says "the code set is the limiting factor"; NOS says "the documentation is the limiting factor." Whenever you find yourself on an NOS code, it is worth asking whether the record supports something more specific before you finalize.
Etiology and Manifestation: "Code First" and "Use Additional Code"
Many conditions have both an underlying cause (etiology) and a resulting effect on a body system (manifestation). ICD-10-CM enforces the order in which these are reported through paired instructional notes.
Code First
A "Code first" note tells you that the underlying condition must be sequenced before the code the note sits under. It appears on manifestation codes — those whose titles often include the phrase "in diseases classified elsewhere" — which may never be reported first or alone. Take dementia in Parkinson's disease: the manifestation code F02.80 ("Dementia in other diseases classified elsewhere, without behavioral disturbance") carries a "Code first" instruction to list the underlying physiological condition first, so the pair is sequenced G20 (Parkinson's disease) followed by F02.80. Sequencing them the other way around is a coding error even though both codes are individually valid.
Use Additional Code
A "Use additional code" note is the mirror image: it appears on the etiology code and tells you to add a secondary code to capture an associated manifestation or, often, an infectious agent. Urinary tract infection illustrates it well: N39.0 ("Urinary tract infection, site not specified") carries a "Use additional code" instruction to identify the responsible organism, so a UTI caused by Escherichia coli is reported as N39.0 followed by B96.20. A "Code also" note is looser still: it tells you two codes may be needed to fully describe the condition but does not mandate the sequence, leaving order to the circumstances of the encounter.
These paired notes are how the classification reconstructs a full clinical story from two atomic codes. When you see one, treat it as non-negotiable: the note is part of the code's definition, not a suggestion.
Brackets, Colons, and the Symbol Conventions
The etiology/manifestation relationship also surfaces as a typographic convention in the Alphabetic Index. There, the manifestation code appears in brackets immediately after the etiology code — for example an Index entry reading G20 [F02.80]. Brackets in the Index always identify manifestation codes that must be sequenced second; in the Tabular List the same brackets enclose synonyms, alternative wording, or explanatory phrases.
A colon is used in the Tabular List after an incomplete term that needs one or more of the modifiers listed beneath it before it can be assigned to a category. The term before the colon is not codable on its own; it becomes valid only when combined with one of the indented modifier terms that follow.
The "And," "With," and Default-Code Conventions
Three word-level conventions carry outsized weight because they change which code is correct without changing a single character in the documentation.
- "And" means "and/or." When "and" appears in a code title, read it as "and/or." A category titled "tuberculosis of bones and joints" covers tuberculosis of the bones alone, of the joints alone, or of both. Reading "and" as strictly conjunctive narrows the code incorrectly.
- The "with" convention. The words "with" and "in" — whether in a code title, an Alphabetic Index subterm, or an instructional note — are interpreted to mean "associated with" or "due to." The classification presumes a causal link between the two conditions even when the provider did not explicitly state one. Under this rule, "diabetes with chronic kidney disease" is coded as a combination code presuming the relationship (for example
E11.22), unless the documentation clearly attributes the kidney disease to a different cause. - Default codes. A code listed immediately next to a main term in the Alphabetic Index is the default code. It represents the condition most commonly associated with that main term, or the unspecified code for it. When documentation offers no further detail, the default is the correct starting point — "pneumonia" with no qualifier defaults to
J18.9.
Putting the Conventions to Work
None of these rules operate in isolation. A single injury code can require you to apply the seventh character, insert placeholder Xs, and honor a "Code first" note for an associated condition all at once. The reliable workflow is to start from the Alphabetic Index to find the default code and any bracketed manifestation, then confirm the full code in the Tabular List, where the seventh-character requirements, placeholder positions, and instructional notes are spelled out. A fast lookup tool compresses that round trip: our ICD-10 code browser shows the complete valid code, its billable status, its parent chain, and the Excludes and instructional notes attached to it, all searchable locally in your browser. For a step-by-step lookup routine, see our guide on how to look up an ICD-10-CM code, and for the bigger picture of what the code set is and how it is maintained, read understanding ICD-10 codes.
Conventions are the grammar of ICD-10-CM. Memorize the handful covered here — the seventh character and its three core episode values, the structural placeholder X, NEC versus NOS, the "code first"/"use additional code" pair, and the "and"/"with"/default rules — and the vast majority of everyday coding decisions become mechanical rather than mysterious. When a case falls outside them, the answer is almost always in Section I.A of the Official Guidelines, which remain the authoritative source no tool replaces.
Sources
- CMS — ICD-10-CM Official Guidelines for Coding and Reporting, FY 2026 — Section I.A defines the conventions covered here, including the seventh character, the placeholder X, and the "code first" / "use additional code" sequencing rules.
- CDC/NCHS — ICD-10-CM — the official U.S. diagnosis classification, developed and maintained jointly by NCHS and CMS.
- CDC/NCHS — ICD-10-CM Files — the annual release files (Tabular List and Alphabetic Index) in which the conventions and instructional notes appear.