A rejection is not a denial — and the difference is money

Most practices track a single number for “claims that did not get paid”. It is the most expensive simplification in the revenue cycle, because the two things inside that number have different causes, different fixes and different deadlines.

What actually separates them

A rejection never entered the payer’s adjudication system. It failed a format or eligibility check at the clearinghouse or the front end, and was handed back before anyone assessed the clinical or contractual merits. In most cases it can be corrected and resubmitted as though it were new.

A denial was adjudicated. The payer received it, considered it against the plan and the contract, and refused it. It cannot simply be resubmitted; it has to be corrected and rebilled, appealed, or closed.

A rejection is a claim that never arrived. A denial is a claim that arrived and lost.

Why combining them hides both

When the two are reported as one figure, the shape of the problem disappears. A month that looks stable can be a month where rejections fell and denials rose — which means the front end improved while something downstream got worse. Averaged together, that reads as no change at all.

• Rejections point upstream: eligibility, demographics, payer IDs, formatting.
• Denials point at coding, documentation, authorisation or the contract.
• Rejections usually have no appeal clock. Denials always do.
• Rejection fixes are cheap and fast. Appeals are neither.

The practical consequence

Reworking is not the same as fixing. Correcting a rejection recovers one claim; identifying that a payer ID has been wrong on every claim from one referring office recovers all of the future ones. That distinction only becomes visible once the two categories are separated and each is coded to a root cause.

If you take one thing from this: split the number. Everything else follows from being able to see which of the two you actually have.