RCM Protocol: Exploit Payer Weakness

This newsletter is for the people in the trenches: the billers, coders and RCM directors reading EOBs and appeal letters every day, up to their necks in the claims process. What I’m sharing here is something you can use today with your existing clients.

Most people look at an abandoned claim as lost revenue. I look at it as an opportunity, and I’d encourage you to as well.

The way most denials get fought

When a claim gets denied, the standard response is to build the strongest possible case for the provider. More clinical documentation. A stronger letter of medical necessity. Proof the prior authorization was submitted correctly. Evidence the coding was right.

That is the documentation argument. It works on a lot of denials. But there’s a category of denials where it will never work. Not because the provider did anything wrong, but because the denial isn’t really about the provider at all. It’s about what the payer’s own systems said.

The part nobody talks about

Payers don’t run on one system. They run on several, usually five and sometimes as many as nine, and those systems rarely communicate. On a technical level there’s no open API connecting them. The EDI gateway, the enrollment registry, the prior authorization crosswalk module and the rest all operate separately.

The silence is intentional, and it is profitable. A system where authorization and adjudication never compare notes is a system that can deny what it already approved. That silence is where the headaches in the claims cycle come from.

The authorization platform that approves a drug is a different system from the claims platform that processes the bill. The clinical policy system that determines coverage updates on a different timeline than the fee schedule system that prices the code. The utilization review department that issues an uphold letter operates independently from the claims department that issued the original denial.

Each of these systems produces its own output. Each output is a document in the correspondence record. And because these systems don’t communicate, they routinely produce outputs that contradict each other. That’s not an occasional problem. It’s a structural one.

Which means most denied claims contain at least one contradiction somewhere in the payer’s own correspondence: a position the payer took in one document that is directly contradicted by a position it took in another.

You don’t go into a denied claim looking for a specific contradiction. You read the full correspondence record as a system and the contradictions surface. Different files produce different contradictions because different documents are present and different systems were involved. Your job is to read every document the payer produced against every other document the payer produced.

Most RCM and billing professionals don’t find contradictions because they read each document as a response to the previous one. The denial arrives and they respond to it. The uphold arrives and they respond to it. That’s not the same as reading the full correspondence as a system, and it’s the difference between a documentation argument and a contradiction argument.

What surfaces when you read the full file

The following examples are not a checklist. They are illustrations of what was found across six different types of correspondence when the full file was read as a system. Each one is invisible if you read documents individually, and visible the moment you read them against each other.

The authorization record vs. the denial letter. The auto-approval paradox: CO-16 / M123

What the payer did: Denied a high-cost claim after the fact, citing insufficient clinical documentation.

Where the contradiction is: Open the electronic 278 prior authorization approval transaction. The metadata shows the authorization was set to Approved with zero required document attachments and no pending clinical reviews flagged at the time of approval.

The argument: The payer’s own approval engine established that the patient qualified and the service was medically necessary based purely on structured data. No clinical documents were required or reviewed. The claims engine cannot retrospectively call documentation insufficient when its own system waived that requirement weeks earlier. The authorization is the clinical determination; the denial contradicts it. The appeal argues the 278 record against the denial, not the provider’s chart.

The authorization database vs. the claims engine. The ghost authorization: CO-197

What the payer did: Denied the claim stating no valid authorization on file.

Where the contradiction is: Pull the payer’s own portal authorization records. The electronic 278 approval shows a signed approval matching the patient ID, CPT/HCPCS code and physician NPI, issued before the date of service.

The argument: The payer’s authorization database contains a valid approval. Its claims engine denied the same claim without checking it. Two systems run by the same payer produced directly contradictory outputs for the same patient, code and NPI. The appeal presents the 278 approval directly against the CO-197 denial. Nothing from the provider’s file is required.

EOB line vs. EOB line. The contracted rate contradiction: CO-45 / CO-96

What the payer did: Denied the drug or procedure line as non-contracted or having no fee schedule entry, while processing and paying other lines on the same claim.

Where the contradiction is: Read the EOB line by line. CO-45 applied to the administration line. CO-96 applied to the drug line. Same claim, same date, same NPI.

The argument: CO-45 is a contracted rate reduction. It applies only when an established contracted fee schedule exists between payer and provider. CO-45 is the payer’s own confirmation that a contract exists; CO-96 is its simultaneous claim that it doesn’t. The contradiction requires no provider documentation, only the EOB.

The denial letter vs. the appeal correspondence. The shifting rationale

What the payer did: Issued an original denial on one basis, upheld the appeal on a different basis and, in some cases, upheld a second appeal on a third basis entirely.

Where the contradiction is: Read the original denial, every appeal acknowledgment and every uphold letter as a sequence, and document the stated rationale at each stage. Original denial: no authorization on file. First uphold: authorization found but expired. Second uphold: medical necessity not established.

The argument: A payer that cannot keep a consistent denial basis across its own written record cannot sustain any of those positions at external review. Each shift is evidence that the original basis could not withstand scrutiny and was quietly replaced. The appeal documents every rationale in sequence and shows they are mutually inconsistent. The payer’s own correspondence is the exhibit.

The 835 remittance vs. the signed contract. The surgical unbundling trap: CO-97

What the payer did: Applied a CCI edit to bundle custom implantable anchors, bone grafts or pass-through hardware into the primary surgical code, setting hardware reimbursement to $0.00.

Where the contradiction is: Pull the signed commercial contract and its pass-through device rider, and read it against the 835 remittance line by line.

The argument: The rider states that specific device lines are excluded from global surgical packaging and must be reimbursed independently at invoice cost. The payer’s claims engine applied a generic CCI edit that doesn’t account for the negotiated terms, and a generic system edit cannot override a signed contractual provision. The appeal is breach of the payer’s own agreement, not a coding dispute.

The 835 remittance vs. the clinical record. The sequential hour deletion: CO-16

What the payer did: Processed a continuous multi-hour infusion billed with an initial administration code and several sequential hour codes. The automated scrubber hit an unprogrammed daily limit and dropped the extended hours as duplicates.

Where the contradiction is: Cross-reference the 835 lines against the EHR nursing time logs and the medication administration record. The remittance shows hours three, four and five dropped. The MAR shows timestamped continuous administration for the full duration.

The argument: The timestamped record confirms the drug was administered for the full billed duration, and the claim reflected that accurately. The payer’s scrubber made a software error, treating sequential time increments as duplicate entries. That is a technical error, not a clinical determination.

The payer didn’t keep this money. Your team left it.

Three questions on every live denial before it goes to appeal

1. Did the payer’s authorization system and claims system produce consistent outputs for this claim?

2. Did the payer keep a consistent denial rationale across every piece of correspondence so far?

3. Does the adjudication on one line of the EOB contradict the adjudication on another line of the same EOB?

If the answer to any of these is no, the claim has a contradiction argument. That argument belongs in the appeal, not the documentation file.

There is no master list

The six examples above came from six different document types across six different denial scenarios. None were visible reading documents individually. All of them were visible the moment the full file was read as a system.

There is no master list of contradictions to memorize. The correspondence is the list. Every file is different because every combination of payer systems, document types and adjudication outputs is different. Read every document the payer produced against every other document it produced, as a system. The contradictions are in there, and that’s where the argument is.

Every example above cost the practice nothing to argue. The payer built the case. The record just needed to be read correctly.

One thing worth fixing before your next file

None of this works on a partial file. You cannot compare an authorization to an EOB if you never received the authorization record. If you run your own shop, make the full correspondence record a condition of taking on a client. If you work inside a platform that only feeds you part of what the payer sent, raise it internally.

These comparisons are not optional extras. They are what separates a real recovery argument from a resubmission that repeats the same fight the payer already won.