ICD-10-CM Structure and Conventions
Key Takeaways
- ICD-10-CM diagnosis coding always starts in the Alphabetic Index and is verified in the Tabular List; never code straight from the Index.
- Conventions such as Includes, Excludes1, Excludes2, Code first, Use additional code, and "in diseases classified elsewhere" control code assignment and sequencing.
- Placeholder X preserves character position when a 7th character or a later character is required on a code shorter than seven positions.
- NOS (not otherwise specified) signals unspecified documentation; NEC (not elsewhere classified) signals specific documentation with no more precise code available.
Why This Chapter Dominates the Score
The CCA (Certified Coding Associate) exam, administered by AHIMA (American Health Information Management Association), contains 105 questions (90 scored plus 15 unscored pretest items), allows 2 hours, and is pass/fail at a scaled score of 300. The largest scored domain is Clinical Classification Systems, so mastery of ICD-10-CM conventions has outsized point value. The current edition tested is FY 2026 ICD-10-CM, effective October 1, 2025, which added 487 new codes, revised 38, and deleted 28.
Structure of a Code
ICD-10-CM classifies diagnoses, reasons for encounters, external causes, and factors influencing health status (Z codes). Codes are 3 to 7 alphanumeric characters. The first character is a letter; the second is numeric; characters 3-7 may be letters or numbers. The first three characters form the category. Characters 4-6 add etiology, anatomic site, severity, and laterality. A required 7th character usually conveys episode of care or sequela.
| Code part | Example in S52.521A | Meaning |
|---|---|---|
| Category (1-3) | S52 | Fracture of forearm |
| Etiology/site/severity (4-6) | .521 | Torus fracture, lower end of right radius |
| 7th character | A | Initial encounter, closed fracture |
The Mandatory Workflow: Index First, Tabular Second
Never assign a code from the Alphabetic Index alone. The Index points to a candidate code; the Tabular List confirms the full character count, instructional notes, Excludes notes, sequencing directions, and any required 7th character. CCA scenarios frequently bury the answer in a Tabular note the candidate would miss by stopping at the Index. A second structural reminder: the Index has two parts an examinee must keep straight — the Index to Diseases and Injuries and the separate Table of Neoplasms, Table of Drugs and Chemicals, and Index to External Causes.
A neoplasm question, for instance, is solved in the Neoplasm Table by reading across the malignant-primary, malignant-secondary, in-situ, benign, uncertain-behavior, and unspecified columns, then verifying in the Tabular.
Convention Signals You Must Recognize
| Convention | What it forces the coder to do |
|---|---|
| Includes | Lists conditions covered by that category. |
| Excludes1 | "Not coded here" — the two conditions cannot be reported together for the same condition. |
| Excludes2 | "Not included here" — the condition is separate and may be coded additionally when present. |
| Code first | Sequence an underlying/etiology code before this one. |
| Use additional code | A second code is needed to fully describe the case. |
| Code also | Two codes may be needed; sequencing depends on the circumstances of the encounter. |
| In diseases classified elsewhere | A manifestation code that can never be principal or first-listed. |
Note the Excludes1 exception added in recent guideline cycles: when the two conditions are clearly unrelated to each other, both may be reported despite an Excludes1 note (for example, an Excludes1 between two conditions that the patient happens to have for genuinely separate reasons). CCA items occasionally test this nuance, so read whether the conditions share a common cause before applying Excludes1 rigidly.
Punctuation, Placeholders, and Abbreviations
Punctuation carries rules. Parentheses ( ) enclose nonessential modifiers that do not change code selection. Square brackets [ ] in the Tabular enclose synonyms, alternative wording, or explanatory phrases; in the Index they identify a manifestation code that must be sequenced second. A colon after an incomplete term means a modifier from the indented list is required to make the term codable. The default convention also matters: when the Tabular lists a code with a default following "see" or with the word in the Index unmodified, the coder uses the default unless documentation supports something more specific.
The placeholder X holds an empty character position so a required later character lands in the correct slot. In T36.0X1A (poisoning by penicillins, accidental [unintentional], initial encounter) the X fills the unused 5th position so the 6th-character (1 = accidental) and 7th-character (A) values are valid. Dropping the X — coding T36.01A — makes the code invalid and unbillable, and a claim with an invalid code is denied. Many injury, poisoning, and toxic-effect categories require the X.
Distinguish the two key abbreviations precisely. NOS (not otherwise specified) equals "unspecified" and signals that the record lacks the detail needed for a more precise code; it is the equivalent of an unspecified code. NEC (not elsewhere classified) signals that the documentation is specific but ICD-10-CM offers no more precise code for that documented condition. Choosing an NEC code when an NOS situation exists — or assigning an unspecified code when documentation supports a specific NEC entry — is a classic CCA distractor.
Code-First / Use-Additional Pairs at a Glance
- A Code first note under a manifestation code ("in diseases classified elsewhere") forces the etiology code ahead of it.
- A Use additional code note adds detail such as an infectious organism, a CKD stage, or a tobacco-use code.
- When both appear, the underlying condition is sequenced first, the manifestation second, and any use-additional codes follow the manifestation. Reading these one-line notes is worth several scored points across the diagnosis section.
A coder finds a diagnosis term in the ICD-10-CM Alphabetic Index. What must the coder do before assigning the final diagnosis code?
An Excludes2 note appears under a category the coder is using, and the excluded condition is also documented for the patient. What is the correct interpretation?
Why would ICD-10-CM require placeholder X in a diagnosis code such as T36.0X1A?