One of the first things that surprises beginner coders about ICD-10-CM is how much information a single code can contain. Unlike a simple number, an ICD-10-CM code is a carefully structured alphanumeric string where every character has a specific meaning. Once you understand how codes are built — character by character — navigating the ICD-10-CM codebook becomes far more intuitive and efficient.
The Basic Structure — Up to Seven Characters
Every ICD-10-CM code begins with a letter and can be up to seven characters long. Each character position carries specific meaning, and codes become more specific as more characters are added. Here is what each position represents:
The example above — S52.001A — means: fracture of upper end of ulna, torus fracture, right arm, initial encounter. Every character adds a layer of specificity that would have been impossible in ICD-9-CM.
Character 1 — The Category Letter
The first character of every ICD-10-CM code is always a letter. This letter identifies the chapter and broad category of the condition. For example, all codes beginning with C are neoplasms, all codes beginning with J are respiratory diseases, and all codes beginning with S are injuries. The letter also indicates which section of the Tabular List you are working in.
Characters 1–3 — The Category
The first three characters together form the category — the broadest level of classification for a diagnosis. Categories represent a single disease, condition, or group of related conditions. For example, J06 is the category for acute upper respiratory infections of multiple and unspecified sites. You should never report a three-character category code if more specific subcategory codes are available — always code to the highest level of specificity.
Characters 4–6 — Subcategory and Code
Characters 4, 5, and 6 add specificity by describing the exact site, etiology, manifestation, severity, or other clinical detail. A decimal point is always placed after the third character for readability — the decimal is part of the code format but is not actually a character in the code structure.
For example, within category E11 (Type 2 diabetes mellitus), the subcategories specify complications — E11.2 is with kidney complications, E11.3 is with ophthalmic complications, E11.4 is with neurological complications, and so on. Further digits specify exactly what type of complication.
The 7th Character — Encounter Type and Episode of Care
Some ICD-10-CM codes require a 7th character to identify the type of encounter or episode of care. The 7th character is mandatory when it exists — a code without the required 7th character is considered invalid and will be rejected by payers. The most common 7th character sets are used in injury and fracture codes (Chapter 19) and obstetric codes (Chapter 15).
Injury and Fracture 7th Characters
| 7th Character | Meaning | When to Use |
|---|---|---|
| A | Initial encounter | Patient is receiving active treatment for the injury — whether at first visit or later visits while still in active treatment phase |
| D | Subsequent encounter | Patient has completed active treatment and is now in the healing or recovery phase — routine follow-up visits |
| S | Sequela | Patient has a late effect or complication that arises as a direct result of a previous injury |
The Placeholder X
Some ICD-10-CM codes require a placeholder X to fill character positions when a 7th character is needed but the code does not have enough characters to reach position 7. The X holds the position so the 7th character lands in the correct spot. Omitting the placeholder X makes the code invalid.
Example: T14.91XA — Suicide attempt, initial encounter. The X fills character 6 so the 7th character A can be placed correctly. Without the X this code would be invalid.
Laterality — Left, Right, and Bilateral
One of the most significant improvements of ICD-10-CM over ICD-9 is the inclusion of laterality in codes. Many codes include a character specifying which side of the body is affected. Common laterality designators are 1 for right, 2 for left, and 3 for bilateral. When laterality is documented, you must capture it — coding to an unspecified side when the documentation specifies left or right is considered under-coding.