What Does an NCCI Correct Coding Modifier Indicator of 0, 1 or 9 Mean?

Finding that two codes are subject to a procedure-to-procedure edit is only half an answer. The edit tells you CMS does not ordinarily expect both codes on the same date of service for the same beneficiary. The Correct Coding Modifier Indicator published alongside it tells you whether there is any route to reporting them separately at all.

Two edits that look identical on the page can behave in opposite ways. One admits no exception under any circumstances. The other admits one whenever the encounter supports it. The pair of codes does not tell you which; the indicator does.

The direction of the pair matters too. CMS lists one code as Column 1 and the other as Column 2, and the Column 2 code is the one denied when both are reported without an appropriate modifier. CMS may publish a pair in one direction, in both, or in neither.

1. Identify which of the two codes CMS lists as Column 2. That is the code at risk, and it is not always the one you expected.
2. Read the indicator on that row. It belongs to the row, not to either code on its own.
3. If the indicator is 0, stop. No modifier changes the result, and appending one does not make the pair reportable.
4. If the indicator is 1, ask the documentation question: were these genuinely separate services under the applicable coding circumstances? CMS recognises a HCPCS Level II subset for this — XE for a separate encounter, XS for a separate structure, XP for a separate practitioner and XU for a separate service.
5. Check the date. An edit row carries its own effective date and, where CMS ended it, a deletion date. A row that does not cover your date of service does not govern your claim.

Indicator 0. A pair is flagged and the row carries an indicator of 0. The correct action is to report the Column 1 code alone, or to review whether the Column 2 service was in fact performed and documented as a distinct procedure that should have been coded differently from the outset. Appending a modifier here is not a remedy; it is a claim CMS has said in advance it does not accept.

Indicator 1. The same pair carries an indicator of 1. Now the question is factual: did the two services occur at a separate encounter, on a separate structure, by a separate practitioner, or as a genuinely separate service? If the record answers yes, separate reporting may be supportable. If the record is silent, the answer is a query to the provider, not a modifier.

Indicator 9. The pair appears in the file but CMS has marked the edit not applicable. Nothing about the pairing is being questioned by this edit.

• Treating the indicator as a property of a code. It belongs to the edit row. The same code can appear in many rows with different indicators.
• Appending a modifier to an indicator-0 edit. The indicator has already said no exception exists.
• Reading an indicator of 1 as permission. It permits separate reporting where the circumstances support it. It does not assert that they do.
• Ignoring the care setting. CMS publishes separate edit files by programme and setting, and a pair edited in one is not necessarily edited in another.
• Ignoring the date. A deleted edit row is not a current edit, and a current row did not necessarily apply on an older date of service.

An indicator of 1 turns the question over to the record. What supports separate reporting is a documented distinction: a different encounter, a different anatomic structure, a different practitioner, or a service that is genuinely separate rather than a component of the other. What does not support it is the presence of the edit, the existence of the indicator, or the fact that both services were performed.

Where the record does not make the distinction, the answer is a provider query. Choosing a modifier to fit the edit, rather than to describe what happened, inverts the reasoning.