CO-97 denial code: is a bundling denial appealable?

CO-97 is one of the few CARCs in our dataset rated a medium, rather than low, appeal-worthiness, and that single fact separates it from most other denials on this list. Whether it is worth contesting depends entirely on why the payer considered the service bundled, which the remittance itself rarely explains in enough detail to act on.

What CO-97 is actually saying

CO-97, published by X12 as "The benefit for this service is included in the payment/allowance for another service/procedure that has already been adjudicated", tells you the payer considers the billed code a component of, or incidental to, a different service already paid on the same claim. It does not by itself say which of three common mechanisms is behind that decision, and the correct response is different for each one.

Three mechanisms, three different fixes

• An NCCI procedure-to-procedure edit. CMS's own bundling table pairs two codes and denies the component code when both are billed on the same date of service. Whether a modifier can fix this depends entirely on the modifier indicator attached to that specific pair, not on whether a modifier feels appropriate.
• A global surgical period. Many E/M visits and follow-up procedures within a defined post-operative window are considered part of the original surgery's payment and are not separately billable, unless the encounter is genuinely unrelated, which modifiers 24 (unrelated E/M) or 79 (unrelated procedure) are built to indicate.
• A missing primary code for an add-on code. Some CPT codes are only valid when billed alongside a specific base procedure. Billing the add-on alone, or without its base code on the same claim, produces a CO-97 that a modifier cannot fix, since the problem is a missing line, not a missing modifier.

Checking NCCI before you touch a modifier

Run the two codes through the NCCI PTP edits before appending anything. If a pair exists and the modifier indicator is 0, no modifier, including 59, changes the outcome; the only correct response is to remove the component code or bill it on a separate date of service. If the indicator is 1, and the documentation genuinely supports the services being distinct, modifier 59 or one of the more specific X modifiers may legitimately bypass the edit. Guessing without checking the indicator is the single most common way a CO-97 appeal fails on a technicality that had nothing to do with the underlying clinical argument.

A worked example: the global period trap

A patient returns three days after a minor outpatient procedure with a new, unrelated complaint at a different anatomic site. The visit is billed as a standard E/M code and denies CO-97, because the practice is still inside the procedure's global period. If the new complaint was genuinely unrelated to the original procedure, and that is documented clearly in the note, modifier 24 correctly signals that to the payer and the appeal has real grounds. If the complaint was actually a follow-up on the same surgical site, the denial is correct as issued, and the fix is not a modifier, it is understanding that routine post-operative care is not separately billable at all.

Once you decide it is worth appealing

A CO-97 appeal only survives if it names the specific mechanism and the specific modifier the documentation supports, the same discipline our guide to writing a reconsideration request sets out in detail. "We believe this was bundled in error" gives a reviewer nothing to check; "the two procedures were performed at different anatomic sites, supporting modifier XS, and the documentation is enclosed" gives them a specific rule and a specific fact to verify.

Frequently asked questions

Is CO-97 always about NCCI bundling?

No. It is also used for global surgical period restrictions and for add-on codes billed without their required primary code. All three produce the same CARC, so the paired remark code and your own review of the two billed codes are what actually tell you which mechanism applies.

Which modifier fixes a CO-97 denial?

There is no single answer. For an NCCI PTP edit with modifier indicator 1, 59 or the more specific XE, XS, XP or XU may apply. For a global period issue, 24 or 79 may apply. For a missing base code, no modifier resolves it; the missing line has to be added instead.

How do I know if my CO-97 denial is worth appealing?

Check whether the service was genuinely distinct from, or unrelated to, the other billed procedure, and whether that is documented in the medical record. If both are true, the appeal has real grounds. If the two services genuinely were part of the same episode of care, the denial is correctly applied and no appeal will overturn it.

Does CO-97 mean the claim itself was billed incorrectly?

Not necessarily. Unlike CO-16, which flags a data error, CO-97 is a payment-methodology decision: the payer is saying it will not pay for this service separately, not that the claim data was wrong. That distinction is exactly why CO-97 sits at medium appeal-worthiness while CO-16 sits at low.