What Is a CPT Code? Categories, Claims, and CPT vs ICD-10 vs HCPCS

Quick answer: What is a CPT code? A CPT code is a standard code that describes a medical service or procedure. It describes what was done for a patient, rather than why the patient needed care. CPT stands for Current Procedural Terminology, and the code set is maintained by the American Medical Association (AMA). Clinicians, coders and payers use the same CPT language so a visit, a test, a procedure or a treatment can be reported, understood and paid in a consistent way.

What CPT actually describes

Think of a claim as a short, structured story with three different kinds of information:

• What was done, the service or procedure. That is the CPT code’s job.
• Why it was done, the diagnosis or reason for the visit. That is the ICD-10 code’s job.
• What else was supplied, for example, a product, a supply or a piece of equipment that CPT does not describe. That is where HCPCS Level II codes come in.

A CPT code on its own does not prove that a service should be paid. Payment also depends on the diagnosis that supports it, the payer’s coverage rules, the documentation in the clinical notes, and whether any required authorisation was in place.

The six Category I sections, explained simply

Most everyday CPT coding uses Category I codes. Category I is organised into six sections, each covering a different kind of care:

Knowing the section is the first step to understanding a code. It tells you what kind of service you are looking at, before you look at the finer detail of what exactly was done.

Category I, Category II and Category III do different jobs

CPT codes also fall into three categories, and the category changes what the code is for:

• Category I codes are the standard codes for established services and procedures. These are the codes most claims use.
• Category II codes are optional tracking codes used mainly for performance measurement and quality reporting, rather than for payment itself. They add extra information about care quality.
• Category III codes are temporary codes for emerging technologies, services and procedures that are still being evaluated. If a temporary code exists for a new service, it is generally used instead of an unlisted code.

When no specific code describes a service at all, an unlisted procedure code from the right section may be used, together with a special report describing what was done. That is an exception, not a shortcut. The report has to explain the service clearly enough for the payer to judge it.

How a CPT code is chosen

Choosing a code is not a guessing game, and it is not done from the bill alone. In a well run workflow it follows the documentation:

1. Start with the clinical notes. What service was actually performed, in what setting, and for how long or with what complexity?
2. Find the right section. Is this a visit, an anaesthesia service, a surgical procedure, imaging, a laboratory test, or another medical service?
3. Match the description to the service. The code description has to fit what was documented, not a similar sounding service, and not a more complex service than the one performed.
4. Check for a modifier. A modifier is extra information added to a code to explain a special circumstance, for example that a service was separate and distinct from another service on the same day, or that it was performed on a particular side of the body. Modifiers are a separate topic and will be covered in their own article; the key point here is that they change how a code is read, not the diagnosis behind it.
5. Check the diagnosis link. Does the diagnosis code support this service? If the reason for the visit does not justify the procedure, the claim is at risk of a medical necessity denial, even if the procedure code itself is correct.
6. Check payer rules before submitting. Some services need prior authorisation, some are bundled with another service, and some are covered only in certain settings. Those rules vary by payer and plan.

A code should never be changed to a different code simply because the first one was denied, unless the documentation genuinely supports the new code. Changing a code without documentation support creates audit and compliance risk.

How a CPT code is used on a claim

Once a claim is submitted, the payer reads the CPT code together with the rest of the claim:

• Adjudication: the payer checks eligibility, coverage, authorisation, coding and the diagnosis and procedure match, then decides what to pay, adjust or deny.
• Fee and bundling rules: the payer’s fee schedule and bundling edits decide whether two services on the same day are paid separately or treated as one combined service.
• The Explanation of Benefits (EOB): the EOB shows the service charged, what was allowed, what was paid, what was adjusted, and any patient responsibility. The CPT description is often what helps a patient recognise which service a line refers to, a topic the next article in this cluster will cover in detail.

That is why the same service can be paid differently by different plans, and why a correct code can still be denied if authorisation, coverage or the diagnosis link is missing.

CPT vs ICD-10 vs HCPCS, at a glance

A common source of confusion is the name HCPCS itself. In the HCPCS system, CPT codes are called HCPCS Level I codes, while the separate Level II set covers the supplies and services above. In everyday billing talk, when someone says “HCPCS code” they usually mean a Level II code, but the two levels are not interchangeable.

ICD-10-PCS is a different set again: it is used for inpatient hospital procedures, not for the professional claims a clinic or practice normally submits. Most readers of this site will meet ICD-10-CM (diagnoses) and CPT (procedures) far more often than ICD-10-PCS.

Common CPT mistakes that lead to denials

Many of the denial reasons in the earlier denials article trace back to coding:

• A code that does not match the documentation, the notes describe one service, the claim reports another.
• A missing modifier, two services are reported as if they were one, or a distinct service is not flagged as distinct.
• A diagnosis that does not support the procedure, the “why” does not justify the “what”, so medical necessity is questioned.
• An unlisted code without a report, the payer cannot judge a service it cannot identify.
• A duplicate code, the same service, for the same patient, on the same date, submitted twice.

Working through those checks before submission catches most preventable coding denials. The full denial checklist is in the denials article linked above.

What this means if you are a patient

You do not need to memorise CPT codes, and you should not try to code your own visit. What is useful to know is:

• The CPT description on an EOB or bill is a good starting point for recognising a service, if you cannot recognise a line, ask the billing office what service it refers to.
• A code on a bill is not, by itself, proof that you owe the full amount. Check the EOB first, then ask whether the charge was denied, adjusted, or is still being corrected or appealed.
• If a service looks wrong, the wrong type of visit, a procedure you did not have, or a duplicate line, raise it with the provider’s billing office with the date of service and the EOB in hand, rather than paying first and disputing later.

Frequently asked questions

Is a CPT code the same as a diagnosis code?

No. A CPT code describes the service or procedure that was performed. A diagnosis code, from ICD-10-CM, describes the condition or reason for the visit. A claim usually needs both, and they have to make sense together.

Who decides which CPT code goes on my claim?

The coding is based on the clinician’s documentation. Coders and billing staff translate that documentation into codes, following CPT guidelines and payer rules. A patient, or a billing office, should not pick a code that the notes do not support.

How often do CPT codes change?

CPT is updated annually, with new codes added and outdated codes revised or removed, and temporary Category II and Category III codes can change during the year. That is one reason a code that worked last year may need checking again this year.

What happens if there is no CPT code for a new treatment?

If a temporary Category III code exists for it, that code is normally used. If no specific or temporary code exists, an unlisted procedure code from the right section may be used with a special report. The payer then judges the service from that report.

Does a CPT code tell me what I will pay?

No, not on its own. What you pay depends on the payer’s allowed amount, your deductible, copay or coinsurance, whether the provider is in network, and whether the service was covered and authorised. The EOB, not the code alone, shows the financial result.

Are CPT, ICD-10 and HCPCS used outside the United States?

CPT is a US system maintained by the AMA and is the standard for professional claims in the US. ICD is an international classification, with the US clinical modification (ICD-10-CM) used for diagnoses in the US. Other countries use their own billing and classification systems, so a code that works in one country does not automatically apply in another.

Sources

• American Medical Association, CPT overview and CPT frequently asked questions
• Centers for Medicare & Medicaid Services, ICD-10 information
• CDC, National Center for Health Statistics, ICD-10-CM information

Cura Editorial Team

Contributor at Cura Health Journal.