Chapter 30 — Appeals: When the Payer Says No and You Know They're Wrong

📍 Where you are

Chapter 29 read Account 10-4471's denial, ran the triage, classified the root cause, and made a decision: appeal. Then it stopped, because deciding to appeal and appealing are different jobs.

This chapter is the second job. What an appeal must contain, how the letter is built, what counts as evidence, and the machinery the appeal travels through — two or so levels at a commercial payer, five at Medicare, and an independent reviewer standing outside both.

The claim in question is correct. Chapter 14 established the modifier, Chapter 21 established the override, Chapter 29 established that the denial is the payer's policy rather than anyone's error. What remains is to demonstrate it, on paper, to a stranger, by a deadline.

Overview

An appeal is the only place in the revenue cycle where you argue.

Everywhere else in this book, the work is compliance: you read the rule and follow it, you find the error and correct it, you see the pattern and prevent it. An appeal is different in kind. It says the payer's decision was wrong, and it says so to the organization that made the decision, on a form that organization designed, inside a window that organization's contract or a federal regulation defines.

Three things decide whether that argument wins, and this chapter is organized around them.

Whether it should have been made at all. Chapter 29 §29.4's ninety-second triage already answers this for most denials, and the answer is usually no. This chapter does not rebuild that test — it stands on it, and adds the three questions the triage does not ask: what is actually at stake, what is the deadline, and which instrument fits. Account 10-5502 opens the chapter for exactly this reason: the most instructive appeal in Part VI is the one that was correctly never written.

Whether the argument is an argument. A payer reviewer reads a stack of appeals, and most of them assert that the service was real, that the doctor ordered it, and that the denial is unfair. None of that is an argument. Chapter 22 §22.6 supplied the reframe this chapter is built on: stop arguing that the service was needed and show that the record contains what the standard asks for. An appeal is a demonstration — the payer's own standard, applied to the record it now has in front of it, produces a different answer than the one its edit produced without the record.

And whether it arrives on time, at the right level, through the right door. The machinery is unforgiving in a specific way: it does not punish weak arguments — weak arguments simply lose. It punishes calendar errors, and those lose appeals that would have won. Chapter 28's Case Study 1 watched two years of winnable appeals expire unwritten; §30.5, §30.6, and §30.10 exist so that never happens on your desk.

The chapter's thesis: an appeal is not a complaint. It is a demonstration, against the payer's own published standard, built from documentation that existed before the claim went out. Chapter 21 §21.11 said it first and this chapter proves it: you do not build it for the appeal. You build it for the claim.

30.1 Deciding whether to appeal at all

Start with the word. An appeal is a formal request that a payer reverse an adjudication decision, made through the payer's or a program's defined review process, arguing that the decision was wrong on the information the payer had or should now consider. That definition carries three requirements that decide everything in this section: there must be a decision (a rejection has nothing to appeal — Chapter 29 §29.1), there must be an argument that it was wrong (not a wish that it were different), and there must be a defined process with its own clock.

Chapter 29 §29.4 already built the merits test and it is not rebuilt here. The integrated triage — Chapter 20's four reasons, Chapter 21's six edit branches, Chapter 22's coded-wrong versus never-going-to-be-covered — takes ninety seconds and most of its branches say do not appeal. If you have not run it, stop and run it. This section begins where the triage ends, with the three questions it does not ask.

Question one: what is actually at stake?

The amount at issue on an appeal is the allowed amount, not the charge. Chapter 28 §28.11 established this on the Encounter itself: line 1 carries a charge of \$185.00, but CO-45 already priced the visit at \$128.40 and CO-97 then removed that allowed amount. **The appealable amount is \$128.40.** An appeal demanding \$185.00 argues against the contract as well as the edit, and the contract argument loses — Chapter 2 §2.6. You are owed what the contract says the service is worth, and that is the number the appeal asks for.

Question two: what is the deadline, and has anything already spent it?

Every appeal right expires, and §30.5 and §30.6 give the windows. The question to ask before writing is whether some earlier action already consumed the right. Chapter 29 §29.6 flagged the trap and this section confirms it: correcting is not appealing. A corrected claim says "here is different information"; an appeal says "your decision was wrong." At some payers, submitting a corrected claim in place of an appeal closes the original determination — the appeal rights attached to it are forfeited, and the corrected claim is adjudicated fresh, sometimes against a fresh timely filing argument. When the claim was right as submitted, do not "fix" it. There is nothing to fix, and the fix costs you the argument.

Question three: is an appeal even the right instrument?

Account 10-5502 — the Harborview MRI — is this book's standing answer. Chapter 22 §22.11 worked it through: the prior authorization for the lumbar MRI (72148) denied with claim adjustment reason code (CARC) 197, the policy requires six weeks of documented conservative therapy, and the record documents four weeks of NSAIDs plus a patient's report of physical therapy. Four is not six, and a report is not documentation. Three rounds of appeal letters cannot change what the record says — an appeal argues about the record, and this record loses. What fixes it is a phone call, four more documented weeks, and a resubmission. The most expensive habit in appeal work is writing letters at problems that are not arguments, and the practice that appealed 10-5502 three times spent three deadlines learning what the triage would have said in ninety seconds: this claim was not coded wrong; this service was never going to be covered, on this record, on this date.

And there is a mirror-image error. Some denials that look unappealable are appealable with the right evidence. CO-29 — timely filing — is Chapter 29's "almost never winnable" category, and the exception is precisely defined: proof that the claim was filed on time. Chapter 27 §27.7 ranked that proof — a payer acknowledgment naming the claim, then a clearinghouse transmission report, then the practice system's history. A CO-29 with a 277CA in hand is not a write-off; it is a short, winnable appeal with one exhibit. The category is not the decision. The evidence is the decision.

🔢 Code It

Six determinations. Appeal or not — and if so, with what?

text 1. CO-97 / N19 on 99214-25; the note documents three chronic conditions managed distinctly from the procedure (commercial payer) 2. CARC 197, prior authorization absent; the policy requires 6 weeks documented conservative therapy; the record shows 4 (Account 10-5502) 3. CO-29, timely filing; the biller finds a 277CA acknowledging the claim eleven days after the date of service (commercial payer) 4. CO-50 on a Medicare claim; the LCD's covered- diagnosis list includes the patient's condition, and the linked diagnosis on the claim is a symptom code instead (Medicare) 5. RARC MA130 — incomplete/invalid claim information 6. CO-45 in the amount of $56.60 on line 1

1 — APPEAL. This is Account 10-4471. Edit denial, modifier indicator 1, documentation supports the override — Chapter 29's triage said appeal, and §30.3 writes it.

2 — DO NOT APPEAL. The record does not contain what the policy requires, and no letter can put it there. The fix is Chapter 22 §22.11's: the call, the four documented weeks, the resubmission.

3 — APPEAL, with one exhibit. The acknowledgment is the ranked proof from Chapter 27 §27.7. This is the exception that proves the timely filing category is about evidence, not about hope.

4 — CORRECT AND RESUBMIT, not appeal. The claim is wrong: the linkage (Chapter 22 §22.6) points at the wrong diagnosis. This is the coded-wrong branch — fixable, and an appeal would be arguing for a claim that misstates the record.

5 — NEITHER. MA130 carries no appeal rights because there is no adjudication to appeal — Chapter 29 §29.5. Fix the claim data and resubmit, or close it.

6 — NOTHING TO APPEAL. CO-45 is the contract working: the spread between the charge and the allowed amount. Appealing it is appealing your own signature. (If the allowed amount is wrong against the contract, that is Chapter 28 §28.8's underpayment method — a dispute, and a different letter.)

The lesson: six determinations, two appeals — and the two appeals have different exhibits. The decision to appeal is a decision about evidence, made claim by claim, after the triage and never instead of it.

The question this section deliberately does not answer

Is a winnable appeal always worth filing? You now hold every input: the allowed amount at stake on any given line, Northgate's published overturn rate — 68%, from Chapter 29 §29.7, and remember that an averaged rate conceals the category that never wins — and the fact that working the Encounter's denial took 58 minutes across three touches (Chapter 29 §29.10). What those numbers assemble into is Q4 — could the denial have been prevented, and what does chasing it actually cost? — and that assembly belongs to Chapter 40 §40.2 and §40.3, which is where this book has sealed it since Chapter 1. This chapter teaches the merits. The economics are the capstone's, and when you reach it you will have filed every number it uses.

⚠️ Where Claims Die

The appeal written because writing one feels like working.

Chapter 21's Case Study 1 is a denials team that spent a year appealing into a category with a zero-percent overturn rate — every letter written by a competent person doing what was asked. The triage exists to prevent exactly that, and this chapter adds the second control: the appeal log (§30.10) records the outcome of every appeal by category, so a category that never wins becomes visible in a quarter instead of a year.

The discipline is symmetrical. Do not appeal what cannot be won — and do not write off what can. The CO-29 with an acknowledgment in hand and the underpaid line with a contract behind it both die quietly in practices that treat "denied" as a verdict instead of as an opening position.

30.2 What an appeal must contain

Strip away every payer's branding and every level's form, and an appeal has six requirements. Miss any one and the appeal fails before anyone reads the argument.

   WHAT EVERY APPEAL MUST CONTAIN

   1  IDENTIFICATION      the claim number, the member, the date of
                          service, the line and code at issue. The
                          payer must be able to find the decision
                          you are appealing.
   2  THE DETERMINATION   what you are appealing: the determination
                          date and the reason code. An appeal of
                          "the denial" of a four-line claim with one
                          denied line makes the reviewer guess.
   3  THE REQUEST         specific and small: "reprocess line 1 and
                          pay at the contracted allowed amount."
                          Not "review this claim."
   4  THE ARGUMENT        why the decision was wrong — argued
                          against the payer's own standard (§30.4).
   5  THE EVIDENCE        the documents that prove the argument,
                          listed and enclosed. Only those documents.
   6  TIMELINESS + CHANNEL  inside the window, at the correct level,
                          through the address or portal the payer
                          designates FOR APPEALS.

Three notes on that list, each of which loses real appeals.

Item 3 is where good arguments go to waste. A reviewer who agrees with you must be able to act in one step. "Reprocess line 1, CPT 99214 with modifier 25, and pay at the contracted allowed amount of \$128.40" can be executed by the person reading it. "We respectfully request reconsideration of this claim" requires the reviewer to figure out what you want, and a reviewer with a stack does not.

Item 6 is two requirements wearing one number. The window is the deadline; the channel is the door. Payers designate appeal addresses and portal queues that are not the claims address, and an appeal mailed to the claims PO box is frequently scanned, matched to the claim, and treated as correspondence — or worse, as a duplicate claim. It does not toll anything. The provider manual names the door; use it, and keep the proof of submission the same way Chapter 27 §27.7 taught you to keep proof of filing.

And the list has a sibling that is not on it: what an appeal is not. It is not a corrected claim (§30.1 — different assertion, different consequence). It is not a records response — when a payer requests documentation before deciding, sending the records is claim processing, not appeal, and the appeal rights have not started. And it is not a grievance, which in most payer vocabularies is a complaint about service or conduct rather than a request to reverse a determination. Payers name these things inconsistently — "reconsideration," "dispute," "redetermination," "review" — and two of those words are Medicare terms of art with fixed meanings (§30.6). Read the denial letter or provider manual for what the process is, not what it is called, and match your filing to the process that carries appeal rights.

🔍 Check Your Understanding

1. A payer's letter asks for the office note "to complete processing" of a claim it has not yet decided. Is your response an appeal?
2. An appeal asks the payer to "review the attached claim for correct payment." What is missing, and what does it cost?
3. Why is the determination date (item 2) load-bearing rather than clerical?

Answers: (1) No — it is a records response during adjudication; no determination exists yet, so no appeal rights have started. Answer it promptly, because CARC 252's clock is running. (2) Items 2 and 3 — which determination, and what specific action. It costs the reviewer's one step, and frequently the appeal. (3) Because every window in §30.5 and §30.6 runs from it — Chapter 28's Case Study 1 lost two years of appeals to windows that ran from determination dates nobody was reading.

30.3 The anatomy of an appeal letter

The appeal letter structure is six parts, in a fixed order, and the order is the argument. Header, ask, standard, demonstration, evidence map, close. A letter built this way can be read in ninety seconds and acted on in one step — which matters, because that is roughly the attention it will get at level one.

   THE SIX-PART APPEAL LETTER

   1  HEADER          claim number · member ID · DOS · line/code ·
                      determination date · reason codes
   2  THE ASK         one sentence. What you want done.
   3  THE STANDARD    the rule the payer's own policy or the
                      governing manual states — quoted or cited,
                      briefly
   4  THE             the documented facts, mapped one by one onto
      DEMONSTRATION   the standard's elements. The longest section
                      and the only one that argues.
   5  THE EVIDENCE    what is enclosed, listed. Nothing enclosed
      MAP             that is not cited; nothing cited that is not
                      enclosed.
   6  THE CLOSE       contact, and the response timeframe the
                      payer's own manual commits to.

Part 3 before part 4 is the entire craft. The instinct is to lead with your facts. Lead with their rule. A reviewer who has just read the standard reads your facts as evidence; a reviewer who reads your facts first reads them as a story, and stories lose to edits. This is Chapter 22 §22.6's reframe operating at the level of paragraph order: the letter does not argue that the service was needed or real — it shows that the record contains what the standard asks for.

Here is the letter for Account 10-4471, written on day 24. It is reproduced in full because the book has been promising it since Chapter 4.

📋 Read the Chart

FIGURE 30.1 — "The level-one appeal" — [Account 10-4471 — constructed teaching example]

```text Northgate Family Medicine April 7 [address · Type 2 NPI · tax ID]

Northfield Mutual Health Plan — PROVIDER APPEALS [appeals address from the provider manual]

RE: LEVEL-ONE APPEAL — request to reprocess one line Member: [member ID] Patient: [name, DOB] Claim number: [payer claim number from the 835] Date of service: March 14 Line at issue: line 1 · 99214, modifier 25 Determination: remittance dated March 31 — CO-97, remark N19, $0.00 paid, allowed $128.40

This is a level-one appeal of the determination above. We ask that line 1 be reprocessed and paid at the contracted allowed amount of $128.40. We are not disputing any other line, and we are not submitting a corrected claim; the claim is correct as filed.

THE STANDARD. Your published payment policy [policy number, title, from the provider portal] states that an evaluation and management service reported with modifier 25 on the same date as a minor procedure is separately payable when documentation supports a significant, separately identifiable E/M service beyond the usual work of the procedure. The NCCI Policy Manual (Chapter I, general correct coding policies) states the same national standard, and states that a different diagnosis is not required. The NCCI procedure-to-procedure edit for this pair carries modifier indicator 1: the edit itself provides for override when documentation supports it.

THE DOCUMENTATION. The enclosed office note (03/14, signed 6:42 p.m.) supports each element:

1. Three chronic conditions — type 2 diabetes, hypertension, hyperlipidemia — are each separately assessed with a separate plan (Assessment items 1-3).
2. Prescription drug management: three medications reviewed and continued.
3. Two laboratory tests (83036, 80061) ordered, each with a stated clinical reason.
4. The knee is a new complaint at this visit, with its own history, its own examination, and a management decision reached at this encounter.

Items 1 through 3 are wholly unrelated to the knee or the injection. They constitute a significant, separately identifiable E/M service on their own, before the new complaint is considered.

On item 4: the note documents that the knee pain is a new complaint at this visit; that no prior imaging of this knee is available; that a focused history and examination were performed and a differential impression recorded; that management options were discussed with the patient at this encounter; and that the patient elected injection the same day, with consent documented in the procedure note. The evaluation that produced the decision is documented at this encounter and nowhere else.

ENCLOSURES. (1) Office note of 03/14 — the only record at issue. (2) Remittance excerpt for this claim, 03/31.

Your provider manual commits to a written level-one appeal determination within 45 calendar days. We have calendared that date. Questions to [billing office contact] at [phone].

[Signature — billing manager] ```

THE DOCUMENT — a level-one appeal letter, one page, submitted day 24 with two enclosures through the payer's appeals portal. THE CONTEXT — the day-17 denial, classified day 20 (Chapter 29 §29.10), appeal decision made, letter written and submitted day 24. Touch 2 of three. WHAT IT SHOWS — the six parts in order. The ask is one sentence and executable. The standard comes before the facts, and it is the payer's own policy first, the national manual second, the edit indicator third. The demonstration maps four documented elements onto the standard and flags that three of them have nothing to do with the knee. WHAT IT DOESN'T — no history of the practice's frustrations, no assertion that the doctor ordered it, no attack on the edit, and no documentation beyond the one note at issue. It also does not quote a sentence saying "the decision to inject was made today," because no such sentence exists — see §30.4 for what it does instead. THE DECISION — submit through the designated appeals channel, keep the submission proof, calendar the payer's own 45-day commitment, and set the practice's 21-day status check (§30.10). THE LESSON — the letter contains nothing that was created for the appeal. Every fact in it was in the note on day 0, and every rule in it was published before the claim was filed. You do not build it for the appeal. You build it for the claim.

Why did this take 31 minutes? Chapter 29 §29.10 recorded touch 2 at 31 minutes and promised this section would show why. Here is where they went: re-read the note against the four elements, 6 minutes. Pull the payer's policy from the portal and confirm the standard's wording, 5 minutes. Pull the Policy Manual language and the edit indicator from the practice's appeal-paragraph library, 3 minutes — the practice keeps its recurring citations as maintained boilerplate, reviewed when the quarterly edit files and the annual Manual revision land, which is what makes minute counts like this possible. Draft the demonstration — the only part written fresh, because it maps this note onto the standard — 10 minutes. Assemble the packet, submit, file the proof, calendar the dates, 7 minutes. Check: 6 + 5 + 3 + 10 + 7 = 31 ✓

The number worth noticing is the 10. Only the demonstration is new work, because only the demonstration is about this patient. A practice that writes every appeal from a blank page is spending an hour where this book spends half of one; a practice that sends the same letter for every denial is sending part 4 empty, and part 4 is the only part that argues.

30.4 Evidence: the note, the policy, the manual, the edit

Four kinds of evidence appear in a coding appeal, and they prove different things. The strongest appeals use all four in one page; the weakest use none of them at length.

The note proves the facts. It is the only document that says what happened at the encounter, and Chapter 4's rule governs everything here: the appeal can use what the provider wrote and nothing else. Enclose the note at issue — and only the note at issue. An appeal packet is a disclosure of protected health information governed by the minimum-necessary standard (Chapter 5 §5.7): the payer reviewing a March 14 denial has no need for the patient's full chart, and sending it is both a privacy failure and a tactical one, because a reviewer handed forty pages reads none of them.

The policy proves the standard the payer must apply. Chapter 22 §22.5 taught you to find it — the payer's own published medical or payment policy, on its provider portal, with a number and a revision date. Citing the payer's own policy converts the appeal from a disagreement into an audit of the payer's compliance with itself, which is the strongest rhetorical position available to a provider. When the policy's criteria are met on the face of the note, say so element by element.

The manual proves the national standard. For edit denials, the National Correct Coding Initiative (NCCI) Policy Manual (Chapter 21 §21.5) is the reference the edit itself is built on. For the Encounter's denial, its general correct-coding chapter does two things at once: it confirms that the routine pre- and post-work of a minor procedure is not separately reportable — which is why the appeal never argues the knee evaluation alone earns the evaluation and management (E/M) service — and it confirms that a significant, separately identifiable E/M service is reportable with modifier 25, with no requirement of a different diagnosis. Quote briefly or paraphrase with a citation; never reproduce pages. And date what you cite: the Manual is revised annually, NCCI edits and HCPCS Level II change quarterly, ICD-10-CM changes every October 1, CPT every January 1 — an appeal citing last year's language invites a response citing this year's.

And the edit proves what the payer's own machinery permits. The procedure-to-procedure edit for the pair at issue carries a modifier indicator (Chapter 21 §21.3), and when that indicator is 1, the edit's own design contemplates override with documentation. Keep Chapter 21's discipline: indicator 1 is a permission, not an authorization. The appeal does not say "the indicator is 1, therefore pay" — it says "the indicator is 1, and here is the documentation the override requires." The indicator opens the door; the note walks through it.

When the record implies but does not state

Here is the hardest and most valuable skill in this section. Sometimes the conclusion you need is true, and supported, and not stated — the note contains every fact that produces the conclusion and no sentence that announces it. Figure 4.2 has exactly this gap, and Chapter 14 §14.4 named it: the note never states that the decision to inject was made during this visit.

The wrong response is to fix the record. The right response is to construct the argument: lay the documented facts in a row and let the reviewer draw the only available conclusion. The letter in §30.3 does it in one paragraph — a new complaint at this visit; no prior imaging of this knee available; a focused history, examination, and differential impression; management options discussed at this encounter; injection elected the same day with consent in the procedure note. Five documented facts, zero invented sentences, one inescapable conclusion: the evaluation that produced the decision happened here. No single quotation could have done that work, because no single sentence in the note says it.

Notice what the construction is for. The predictable counterargument to any modifier-25 appeal is that the visit was really a procedure appointment, and the E/M merely its built-in assessment. The constructed paragraph forecloses it — this was not a scheduled injection; the knee arrived as a question and left as a decision — and the three knee-independent elements then stand on their own. Argue from the record, in this order, and the letter does not depend on any one sentence the provider happened to write.

⚖️ Compliance Check

The line between constructing an argument and constructing a record is bright, and this is the chapter that has to say so.

Constructing an argument — §30.4, above — uses only facts already documented, arranged so their consequence is visible. That is advocacy, and it is the job.

Constructing a record is asking the provider, after a denial, to add the sentence the appeal wishes existed. An amendment or late entry made after a denial, to win the denial, is visible — Chapter 4 §4.5: amendments are dated, attributed, and flagged as amendments in the legal health record — and a reviewer who sees a post-denial addendum asserting exactly the disputed element will discount the note entirely, and may refer the file rather than pay it. Chapter 29's Case Study 2 is the template version of the same disease.

If the record genuinely fails to support the code, the remedy is Chapter 29 §29.6's corrected claim, not a stronger letter — and prospectively, a better template (Chapter 38 §38.3 governs what a query may ask, and the rules do not relax because money is at stake). Requirements vary by state and payer, and your compliance officer, not this book, is the authority for your organization.

The appeal must also be true. An appeal to a federal health program is a statement to the government about a claim — Chapter 5 §5.1's certification logic reaches it, and a pattern of appeals asserting documentation that does not say what the letters claim is a False Claims Act theory, not a persistence strategy. Write every letter as if the reviewer will read the note beside it, because the good ones do.

30.5 Commercial appeal levels and deadlines

There is no single commercial appeal system. There is a contract, a provider manual, state law, and — for self-funded plans — a federal statute, stacked in that order, and the first job on any commercial appeal is finding out which stack you are standing in.

The typical structure

Most commercial payers offer one or two internal levels. (Typical; your contract and manual govern — verify.) Level one is a review inside the payer's claims or appeals unit — frequently the first time a human reads the documentation, which is why level one overturns edit denials at meaningful rates: the edit fired on codes; the appeal supplies the record the edit never saw. Level two is a fresh review by someone not involved in level one, often with a clinical reviewer for medical-judgment questions, and it is usually the final internal step. After the internal levels, the routes divide: external review (§30.9) for medical-necessity and medical-judgment questions, and — for contract questions like pricing — whatever dispute-resolution clause the participation agreement contains.

Timely filing for appeals is the deadline discipline applied to the appeal itself: every level has a filing window, and every window runs from the date of the determination being appealed — level one from the remittance date, level two from the level-one decision letter. Windows of 60 to 180 days from the determination are common for level one (typical range; your contract governs). Two consequences follow. First, the clock is running while the denial sits in a work queue — Chapter 29 §29.5 sorted the queue by deadline for exactly this reason, and the deadline it sorts by is this one. Second, the determination date is a fact you read off the remittance, not a fact you remember — Chapter 28's Case Study 1 lost the great majority of two years of winnable bundling appeals because denials posting as contractual adjustments generated no work items, and the windows ran out unobserved, one remittance at a time.

🧮 Run the Numbers

The Encounter's appeal calendar, in dates. (Constructed — Northfield's provider manual, a Tier-3 document: level-one appeals due within 180 days of the determination; the plan responds within 45 calendar days.)

text Determination (remittance) day 17 Fri Mar 31 Appeal window closes day 17 + 180 days Appeal submitted day 24 Fri Apr 7 ► 7 days used, 173 to spare Practice status check day 45 Fri Apr 28 ► 21 days after submission (§30.10) Payer's own commitment day 24 + 45 = day 69 ► Mon May 22 Decision letter day 59 Fri May 12 ► inside its window Second remittance day 66 Fri May 19

Checks: 24 − 17 = 7 days from determination to filing ✓ · 45 − 24 = 21 days to the status check ✓ · 59 − 24 = 35 days to decision, within the 45-day commitment ✓ · 59 − 17 = 42 days from denial to decision — Chapter 21 §21.9's "forty-two days later" ✓

Two readings. The practice used 7 of 180 days — not because the deadline pressed, but because the claim's value decays while it waits (recurring theme 6), and because a fresh denial is a fresh memory. And both calendars are written down: the payer's commitment is calendared the day the appeal goes out, because a deadline nobody recorded is a deadline nobody can enforce.

The ERISA difference

Chapter 2 §2.5 promised this chapter would cover the practical differences for self-funded plans, and they change more than most billers expect. A self-funded plan is governed by the Employee Retirement Income Security Act (ERISA), federal law, and state insurance law generally does not reach it — not the state's prompt-pay statute, not its appeal-deadline mandates, and frequently not its external review process.

What that means at the desk:

The rights are the member's. ERISA's appeal rights belong to the plan participant, and the Department of Labor's (DOL's) claims-procedure regulation (29 C.F.R. § 2560.503-1) is where they live: the member must get at least 180 days to appeal an adverse benefit determination, the plan must decide within regulated timeframes, and the member is entitled, on request and free of charge, to the documents, guidelines, and criteria the denial relied on — which is how you obtain a self-funded plan's version of the policy §30.4 needs. A provider usually appeals as the member's authorized representative, using the plan's designation procedure; the network contract may add a separate provider dispute channel on top, but the ERISA rights ride with the patient.

The complaint route is federal. When a self-funded plan misses its own deadlines, the state insurance department has no jurisdiction; the Department of Labor does. Cite the regulation, in writing, before anything else — a plan administrator's obligations under §2560.503-1 are specific, and a letter that names them tends to get a response that a phone call does not.

And you frequently cannot tell from the card — Chapter 2 §2.5's advice stands: ask whether the plan is self-funded, and write the answer in the account notes with the date and the representative's name. On an appeal, that one fact chooses the deadline, the standard, the reviewer, and the regulator.

30.6 The five levels of Medicare appeal

Medicare replaces the contract with a statute. The levels of Medicare appeal are five, fixed in law and regulation, identical at every Medicare Administrative Contractor, and published rather than negotiated — which makes them the cleanest appeal system in this book and the one the certification exams test by name.

   THE FIVE LEVELS OF MEDICARE APPEAL  (Part A / Part B claim appeals)

   LEVEL  WHO DECIDES                    FILE WITHIN         DECIDES IN*
   ─────  ─────────────────────────────  ──────────────────  ───────────
     1    REDETERMINATION                120 days from       60 days
          the MAC — a fresh look by      the initial
          the contractor that decided    determination
   ─────  ─────────────────────────────  ──────────────────  ───────────
     2    RECONSIDERATION                180 days from the   60 days
          the QIC — an independent       redetermination
          contractor, on the record      decision
   ─────  ─────────────────────────────  ──────────────────  ───────────
     3    ALJ HEARING                    60 days             90 days*
          an administrative law judge    ► amount in
          at OMHA — the first level        controversy
          outside the claims world         required
   ─────  ─────────────────────────────  ──────────────────  ───────────
     4    MEDICARE APPEALS COUNCIL       60 days             90 days*
          Departmental Appeals Board
   ─────  ─────────────────────────────  ──────────────────  ───────────
     5    JUDICIAL REVIEW                60 days
          federal district court         ► higher amount
                                           in controversy

   * statutory standards; see Case Study 1 for what happened
     when volume overwhelmed them. Amount-in-controversy
     thresholds adjust annually — verify the current figures
     at CMS/OMHA before relying on them.

Walk the ladder once, because each level is a different kind of review.

Level 1 — redetermination — is the Medicare Administrative Contractor (MAC) taking a second look at its own initial determination, by staff not involved in the first decision. Chapter 3 §3.3 listed this among the MAC's jobs and deferred the mechanics here. There is no minimum dollar amount, and for a participating provider on an assigned claim, the right to file is the provider's own. The initial determination arrives on the remittance advice — the beneficiary's copy is the Medicare Summary Notice (Chapter 3 §3.9) — and receipt is presumed five days after the notice date unless shown otherwise.

Level 2 — reconsideration — belongs to a Qualified Independent Contractor (QIC): a contractor whose entire function is second-level review, organizationally independent of the MAC that made and re-made the decision. The review is on the record — no hearing — and it carries the evidence rule that §30.7 calls the most consequential sentence in the Medicare appeal system.

Level 3 — the administrative law judge (ALJ) — moves the appeal out of the claims apparatus entirely, to the Office of Medicare Hearings and Appeals (OMHA), an HHS office separate from CMS. An ALJ hearing is a genuine adjudication — testimony, usually by phone or video, with the provider able to explain the medicine and the coding to a judge. It is also the first level with an amount in controversy requirement, a minimum dollar amount at issue, adjusted annually; smaller claims may be aggregated to reach it. (Verify the current threshold — it changes each calendar year.)

Level 4 — the Medicare Appeals Council — sits within the HHS Departmental Appeals Board and reviews ALJ decisions, on request or on its own motion. Level 5 — judicial review — is a civil action in federal district court, with a higher amount-in-controversy threshold, also adjusted annually. By level 5 you are in litigation, with counsel, and out of this book's territory.

Two structural facts complete the picture. The ladder must be climbed in order — each level reviews the one below, and skipping is not offered. And the deadlines run one way: when a level misses its own decision timeframe, the regulations let the appellant escalate the appeal to the next level rather than wait indefinitely — a right that mattered enormously in the years Case Study 1 describes. (Escalation has its own procedures and trade-offs; verify current mechanics before using it.)

For a working practice, levels 1 and 2 are the whole game. They are free, fast by the standards of what follows, decided on the record you assemble — and they are where a documentation appeal like the Encounter's would be won. The upper levels exist for the big disputes: extrapolated overpayments (Chapter 37 §37.6), pattern denials, statutory questions. Know the whole ladder; work the bottom of it well.

🎓 Exam Watch

The five levels are exactly the kind of named sequence a certification exam tests — the order, the decider at each step, and the filing windows. Billing-oriented credentials lean on it hardest; check your own exam's published content outline for the weight it carries.

The classic trap is the first two levels, because the words sound interchangeable: REDETERMINATION is first and belongs to the MAC. RECONSIDERATION is second and belongs to the QIC. Alphabetical order is the mnemonic that survives exam pressure: rede- before recon-, MAC before QIC.

Second trap: who is independent. The redetermination is the same contractor, different staff; the QIC is a different organization. A stem describing "review by an independent contractor" is level 2.

Third: the windows are not symmetrical — 120, then 180, then 60, 60, 60. And the amount in controversy starts at level 3, not level 1; a question offering a dollar threshold for redetermination is testing whether you know there is none.

30.7 Redetermination and reconsideration in practice

The ladder is the doctrine; this section is the desk. A redetermination is the level-one appeal: a written request that the MAC re-examine its initial determination, filed within 120 days, decided generally within 60 days, with no minimum amount. A reconsideration is the level-two appeal: a written request that the QIC review the redetermination, filed within 180 days of it, decided generally within 60 days, on the record.

Filing a redetermination

The request is simple by design. The MAC's portal is the ordinary channel; the standard paper form is CMS-20027 (form numbers and processes change — use your MAC's current version); and a plain written request works if it identifies the beneficiary, the Medicare number, the specific items and dates of service, and what you disagree with, with a signature. Everything §30.2 requires, in other words — the Medicare forms are §30.2's list turned into fields.

Build it like §30.3's letter anyway. The form gets the appeal into the system; the attachment wins it. A redetermination request whose "reason for disagreement" box says "see attached" and whose attachment is a one-page six-part letter with the note enclosed is the same craft as the commercial appeal, pointed at a different reviewer. The standard you cite is different — for coverage denials it is the NCD or LCD itself (Chapter 22 §22.3–§22.4), and an LCD denial is appealed by showing the record meets the LCD's own criteria, the exact discipline of Chapter 22 §22.6.

And mind who is filing. On assigned claims the provider appeals in its own right. Where the practice appeals on the beneficiary's behalf instead — or a billing agent handles appeals — the appointment of representative (form CMS-1696) is the instrument that makes the filing proper. (Verify current form and scope with your MAC.)

📋 Read the Chart

FIGURE 30.2 — "The redetermination decision" — [constructed teaching example]

```text MEDICARE REDETERMINATION NOTICE [MAC name]

Beneficiary: [name] Medicare number: [MBI] Provider: [practice] Claim: [ICN] Date(s) of service: [date] Decision date: [date]

DECISION: PARTIALLY FAVORABLE

Service 1 [code] originally denied NOW PAYABLE The documentation submitted with your request supports coverage under LCD [number]. The claim will be adjusted; payment will appear on a future remittance advice.

Service 2 [code] originally denied DENIAL UPHELD The record does not document [the LCD's required element]. The information considered is listed below.

WHAT TO DO NEXT: If you disagree, you may request a RECONSIDERATION by a Qualified Independent Contractor within 180 days of the date of this notice. A reconsideration request form is enclosed. IMPORTANT: Submit all evidence you want considered with your reconsideration request. Evidence submitted later may not be considered absent good cause. ```

THE DOCUMENT — the Medicare Redetermination Notice (MRN), the MAC's written level-one decision. THE CONTEXT — a two-service redetermination; one service overturned on the documentation, one upheld. WHAT IT SHOWS — three things every MRN gives you: the decision per service, the reason tied to the governing policy, and the next level with its deadline. The partially favorable outcome is ordinary — appeals are decided line by line, like everything else in adjudication. WHAT IT DOESN'T — money. The favorable line pays through a later remittance and must be reconciled there (Chapter 28 §28.5); an appeal is not resolved until the adjustment posts (§30.10). THE DECISION — post the win when it arrives; for the upheld line, run the triage again against the stated reason. If the record genuinely lacks the required element, a reconsideration cannot supply it, and the honest move is Chapter 29's: classify, close, prevent. THE LESSON — the notice itself tells you the two things the next level needs: the deadline, and the warning this section turns to now.

The reconsideration, and the evidence rule

The QIC's review is on the record — and the record closes there. For providers and suppliers, the regulations require that a reconsideration request present all the evidence the appellant wants considered; evidence introduced later in the ladder can be excluded absent good cause for the late submission. The MRN's warning paragraph is that rule talking.

The practical meaning is blunt: the reconsideration is the last easy chance to complete the file. Whatever you were saving — the second note, the flowsheet, the therapy log that proves the six weeks — goes in now, organized and cited, or it may never be seen by the ALJ you are hoping will finally understand. Assemble the reconsideration as if it were the hearing, because evidentially, it is.

⚠️ Where Claims Die

Holding evidence back is a strategy from television, and in Medicare appeals it is a self-inflicted wound.

The instinct — lead with a little, save the strong material for the judge — inverts the system's actual design. The record substantially closes at level 2. The appellant who arrives at OMHA with a persuasive document the QIC never saw is asking a judge for a good-cause finding before the judge will even look at it.

The discipline that wins instead is front-loading: the complete argument at redetermination, the complete record at reconsideration — nothing new left to say by level 3, only a judge who has not yet heard it. Which is §30.3's craft again: the appeal that wins is complete early, and the Encounter's level-one letter enclosed everything it cited on day 24.

30.8 The peer-to-peer and who should make the call

Chapter 22 §22.10 defined the peer-to-peer review — the treating physician and the payer's physician reviewer, on the phone, about one case — and made the coder's role clear: know it exists, know the deadline, get it scheduled. This section is about the half Chapter 22 left open: who should make the call, and how the call is won or lost.

Who: the treating physician. Not the biller — the payer's medical director will not discuss clinical judgment with a billing office, and should not. Not, if avoidable, a partner covering from the group — the reviewer's questions are about this patient's presentation and this decision, and "I'd have to check the chart" spends the only currency the call has. The physician who saw the patient is the only person who can say what the record cannot, which is exactly the peer-to-peer's value: Chapter 22 noted that a conversation surfaces facts a form never asked for.

When: mostly before or during, not after. The peer-to-peer lives chiefly in the prior authorization and concurrent review world — the Harborview family of denials, inpatient status reviews, continued-stay decisions — where a conversation can still change the determination before it hardens into a claim denial. Windows are short, sometimes days (payer-specific — verify). Some payers offer a post-service equivalent inside the appeal process; many do not, and for a post-service documentation denial like the Encounter's, the written appeal is the instrument — there is no clinical judgment in dispute on Account 10-4471, only whether the record supports the modifier, and that question is decided by reading, not by conversation.

How: preparation, which is the billing office's contribution. The physician makes the call; the revenue cycle makes the call winnable:

   THE PEER-TO-PEER PREP SHEET  (one page, to the physician,
                                 before the call)

   1  THE DECISION AT ISSUE   what was denied, when, and the
                              stated reason — verbatim
   2  THE STANDARD            the payer's criteria for this
                              service, with the two or three
                              elements the reviewer must find
   3  THE FACTS THAT MEET IT  where each element lives in the
                              record — dates, values, failed
                              therapies, findings
   4  THE GAP, NAMED          what the reviewer will say is
                              missing, and the honest answer
   5  LOGISTICS               reference number, the scheduled
                              time, the reviewer's name if
                              known, the callback number

Item 4 is the one practices skip and should not. If the record has a real gap — Harborview's four weeks against a six-week requirement — the physician needs to know before the call, because the reviewer certainly will. A peer-to-peer cannot argue a gap out of existence; what it can do is establish clinical facts that satisfy the criterion another way, or clarify what additional documentation would change the answer, which converts a denial into a plan — Chapter 22 §22.11's outcome, reached by phone.

And afterward: document the call. Date, time, reviewer's name, reference number, outcome, and any commitment made — in the account notes, immediately. A verbal approval without a reference number has a way of not existing when the claim adjudicates.

📞 On the Phone

A peer-to-peer that goes well sounds like a case presentation, not a negotiation.

Physician: "Thanks for the time. This is the 58-year-old with six weeks of medial right knee pain. Your criteria ask for failed conservative therapy and a clinical exam consistent with the indication — she's had six weeks of NSAIDs with only partial relief, documented from [date], and on exam she has medial joint line tenderness and crepitus with a stable ligament exam. That's the basis for the injection."

Reviewer: "The referral we received documents four weeks."

Physician: "The referral summarized it; the office note from [date] documents onset six weeks prior with daily NSAID use since. I can have that note sent today."

That is the call working — a specific criterion, a specific fact, a specific document, and an offer to close the gap in writing.

The failure modes are just as specific. The physician who opens with "this denial is ridiculous" has made it adversarial and clinical facts now arrive as complaints. The physician who was not briefed answers "I'd have to look" three times, and the reviewer — who is also reading from criteria — upholds for lack of information. And the office that wins the call but records no reference number gets to discover, on the remittance, that verbal approvals are only as durable as their paper trail. (Constructed dialogue — a teaching example, not a transcript.)

30.9 External review and the independent reviewer

Every process so far has one structural weakness: the organization reviewing the decision made the decision. External review is the answer. An external review is a review of a plan's final adverse determination by an independent review organization (IRO) — a reviewer outside the payer, accredited for the purpose, whose decision binds the plan.

The right is statutory. The Affordable Care Act (§2719 and its implementing rules) requires non-grandfathered commercial plans to provide external review of denials that turn on medical judgment — medical necessity, appropriateness, level of care, experimental/investigational determinations — and of rescissions. The machinery divides along the line §30.5 drew: insured plans follow their state's external review process where the state runs a conforming one; self-funded ERISA plans use the federal external review process. (Grandfathered plans and some plan types sit outside these rules — verify which regime governs the plan in front of you.)

The practical parameters: (regulatory standards — verify current details)

• Exhaustion first, usually. External review follows the plan's internal appeal process — with exceptions for urgent situations and for plans that fail to follow their own procedures.
• The window is generous by this chapter's standards: four months from the final internal adverse determination to request external review.
• The requester is the member — or you, standing in for the member. Like the ERISA rights it extends, external review belongs to the patient; a provider ordinarily files as the member's authorized representative. The network contract's dispute clause is a separate track and does not substitute.
• The decision binds the plan, which is what makes the four-month window worth calendaring: it is the only reviewer in the commercial stack whose "pay it" the payer cannot decline.

Know what external review is for — and not for. It reviews medical judgment. A pricing dispute, a bundling edit, a timely-filing argument, a contract-interpretation fight — none of these belongs to an IRO, and requesting external review of them wastes the window (state processes vary at the margins — some review a broader class of adverse determinations — but medical judgment is the core). The Encounter's CO-97 never goes here: it was resolved two levels earlier, on documentation, and it is not a medical-judgment question. The denials that do go here are Chapter 22's — necessity, level of care, investigational status — after the internal levels fail.

Medicare is not in this system. Its five levels are the external machinery, built in: the QIC, the ALJ, and the Council are all reviewers outside the MAC. Medicare Advantage runs its own Part C version with a feature worth knowing: when an MA plan upholds its own denial on the plan-level appeal, the case files are forwarded automatically to an independent review entity — the enrollee does not have to ask. (Part C appeal mechanics have their own deadlines and vocabulary — verify current procedures before working an MA appeal.)

🔍 Check Your Understanding

Route each to the right machinery:

1. A fully insured commercial plan upholds, at its final internal level, a denial of an inpatient stay as "not medically necessary at that level of care."
2. A self-funded plan's TPA denies a service as experimental; internal appeals are exhausted.
3. A Medicare Part B claim denies under an LCD; the record appears to meet the LCD's criteria.
4. A commercial payer pays a line at 60% of the contracted rate and the contract is clear.

Answers: (1) State external review — a medical-judgment question on an insured plan, filed within four months, typically as the member's authorized representative. (2) The federal external review process — ERISA plan, medical judgment, state review does not reach it (§30.5). (3) The Medicare ladder: redetermination at the MAC, arguing the LCD's own criteria element by element (§30.7). (4) None of the above — that is not a medical-judgment question or a Medicare appeal; it is an underpayment (Chapter 28 §28.8) worked through the contract's dispute provisions.

30.10 Tracking appeals so they do not expire

An appeal generates two calendars, and practices reliably keep only one of them. Yours: the filing windows at each level, for every open denial. Theirs: the decision deadlines the payer or the regulation commits to — the 45 days Northfield's manual promises, the 60 days the redetermination regulation sets. A practice that tracks only its own deadlines files on time and then waits indefinitely; a practice that tracks both files on time and collects on time.

The instrument is the appeal log — the appeal-stage sibling of Chapter 29 §29.7's denial log, one row per appeal:

   THE APPEAL LOG — one row per appeal

   claim · line · payer · AMOUNT AT ISSUE (allowed, not charge) ·
   root cause category (Ch. 29 §29.4) ·
   LEVEL · date determination received · FILING DEADLINE ·
   date filed · proof of submission ·
   PAYER RESPONSE DUE (their clock) ·
   NEXT FOLLOW-UP (your clock) ·
   status · OUTCOME · date the money posted

Four disciplines make the log work, and each one exists because its absence loses money.

Enter the row when the appeal decision is made, not when the letter is written. The most dangerous interval in appeal work is between "we should appeal this" and "we appealed this" — a claim in that gap is on nobody's calendar while its filing window runs. The Encounter's row was opened on day 20, at classification, with the deadline computed from the day-17 determination; the letter followed on day 24.

Work the log by date, not by dollar. The same rule as Chapter 29 §29.5's queue, for the same reason: expiration is the only irreversible event in this chapter. The daily view is "what expires soonest?" — both directions: filings coming due, and payer responses now overdue.

Follow up on a schedule, not on a memory. The Encounter's practice checks status 21 days after every submission (constructed house rule) — which is what put the day-45 call on the calendar that Chapter 29's touch 3 recorded. The status check is not decoration: it confirms the appeal is in the payer's system as an appeal (§30.2's wrong-door failure surfaces here, while it can still be fixed) and it timestamps the payer's clock in your notes.

And an appeal is not resolved until the money posts. A favorable decision letter is a promise; Chapter 28 §28.5 posts promises against remittances. The outcome field closes when the adjustment appears — the Encounter's on day 66 — and that outcome flows back into Chapter 29 §29.7's log, where it becomes the overturn-rate-by-category that decides which appeals get written next quarter. When the payer's own response deadline passes in silence: escalate in writing — cite the manual's commitment or the regulation, request the determination, and copy the account notes; for insured plans the state insurance department stands behind the deadline, for ERISA plans the Department of Labor's regulation does (§30.5), and for Medicare the escalation right itself does (§30.6).

📋 Read the Chart

FIGURE 30.3 — "The appeal log, exported" — [constructed teaching example]

Source: a practice's open-appeal report, month end, handed over with "appeals are going fine"

```text OPEN APPEALS ......................... 31 FILED THIS MONTH ..................... 9 DECIDED THIS MONTH ................... 6 (4 favorable)

AGING OF OPEN APPEALS (from filing) 0–30 days .......................... 11 31–60 days ......................... 8 61–90 days ......................... 5 over 90 days ....................... 7

ROWS FLAGGED no NEXT FOLLOW-UP date ............. 9 payer response deadline PASSED ..... 6 favorable decision, no posting ..... 3 "pending — level?" ................. 4 ```

What it means: the wins are real and the report is still bad news.

Six appeals sit past the payer's own response deadline, and nothing in the workflow noticed — the payer's calendar is not being tracked, so the payer's commitments cost it nothing.

Nine rows have no next-follow-up date, which means nine appeals are waiting on somebody's memory. The over-90 bucket is where those rows end up.

Three favorable decisions have no posted payment. A won appeal that never posts is a denial with better paperwork — Chapter 28's posting discipline is the last step of this chapter's process, and these three rows are stalled one step from done.

And four rows do not record the level, which means their next filing deadline cannot be computed. Those are the rows that expire.

What to do about it: none of this needs staff. It needs four required fields — level, filing deadline, payer due date, next follow-up — and a daily sort by date. The log is a control only if its dates are filled in; otherwise it is a list of good intentions — Chapter 27's line about reports applies to logs, and to this one specifically.

Where it appears: any practice that measures appeals by how many were filed. Filed is activity. Decided, posted, and on time is performance.

30.11 🗂️ The Encounter — the appeal that won Account 10-4471

Day 24, Friday, April 7. The remittance landed day 17; Chapter 29 classified the denial on day 20 and decided to appeal. Touch 2 — 31 minutes, itemized in §30.3 — produced the letter in Figure 30.1 and submitted it with two enclosures through Northfield's appeals portal. This checkpoint assembles what this chapter contributed to the file: the argument, the evidence, the deadline, and the outcome — and why this one was won.

The argument. The letter never argues that the visit happened or that the doctor thought it necessary. It demonstrates, against Northfield's own published policy and the NCCI Policy Manual's standard, that the record documents a significant, separately identifiable E/M service: the four elements frozen since Chapter 4 — three chronic conditions each separately assessed with a plan, prescription drug management, two labs ordered with stated reasons, and a new problem with its own history, examination, and management decision — three of which have nothing to do with the knee. That last clause is the argument's spine: even if the payer folded every knee-related sentence into the injection, the E/M stands on the other three.

The constructed paragraph. The note's one documented gap — it never states that the decision to inject was made during this visit — was handled the way §30.4 teaches: not by quoting a sentence that does not exist, and not by asking the provider for a post-denial addendum, but by laying five documented facts in a row — new complaint this visit, no prior imaging of the knee available, focused history and exam with a differential impression, options discussed at this encounter, injection elected the same day with documented consent — and letting the reviewer draw the only available conclusion. That paragraph is why this file appears in this chapter: it is what appeal-writing actually is when the record is good but not perfect, which is most of the time.

The evidence. Three documents, exactly as Chapter 6's toolkit table predicted on row 13: the office note, the NCCI Policy Manual's language, and the payer's own published policy. The note was the only clinical record enclosed — minimum necessary — and every rule cited was published before the claim was filed. Nothing in the packet was created for the appeal. You do not build it for the appeal. You build it for the claim — and this file is that sentence, kept.

The deadline. Filed on day 7 of a 180-day window — not because the window pressed, but because a claim's value decays while it sits and the practice's log (§30.10) works by date. Both calendars were written down at submission: the practice's 21-day status check landed on day 45, and Northfield's own 45-day commitment was calendared to day 69 and never needed.

The outcome. The decision letter arrived on day 59 — upheld in the provider's favor, 42 days after the denial. The second remittance posted on day 66: line 1 paid at the full \$128.40 allowed — \$98.40 from the plan, \$30.00 to patient responsibility, the copay finding its home at last (Chapter 28 §28.11). Check: 98.40 + 30.00 = 128.40 ✓. And with it, Q1 closes its last loop: Chapter 14 answered that modifier 25 was correctly applied; a human reviewer at the payer, reading the note, has now agreed.

What this settles. Line 1's money, the last open dollar of the payer's side of this account. The correctness of the claim, now confirmed by the one audience that had disputed it. And the file's demonstration of this chapter's thesis: the appeal won because the documentation, the policy, and the manual said the same thing, and the letter did nothing but put them side by side.

What it does not settle. Whether any of this should have been necessary. The appeal consumed 31 of the denial's 58 minutes and 49 days of calendar (day 17 to day 66) to collect an amount the contract had priced before the patient checked in. Q4 — could the denial have been prevented, and what does the chase actually cost? — now has every input on the record: the minutes (Chapter 29), the overturn rate (Chapter 29 §29.7), the amount (Chapters 2, 23, 28), and, coming next, the cost side of the ledger in Chapter 31 §31.7.

Chapter 40 assembles them. This chapter, like the two before it, stops at the door.

Open questions: Q4 remains open — and it is the only one left.

Summary

An appeal is a demonstration, not a complaint: the payer's own standard, applied to the record it now has, produces a different answer. It requires a decision to appeal (a rejection has none), an argument that the decision was wrong, and a defined process with a clock.

Decide first, and decide with Chapter 29's triage plus three questions it does not ask. The amount at stake is the allowed amount, not the charge — \$128.40, not \$185.00. The deadline may already be spent — correcting is not appealing, and a corrected claim can forfeit the original determination's appeal rights. And the instrument must fit: Account 10-5502's fix was a phone call, four documented weeks, and a resubmission — three appeal letters could not change what the record said. The mirror image is equally real: a CO-29 with a 277CA behind it is a winnable appeal with one exhibit. The category is not the decision; the evidence is.

Every appeal contains six things: identification, the determination being appealed, a specific executable request, an argument against the payer's own standard, the evidence listed and enclosed, and timeliness through the designated channel. The letter's six parts put the standard before the facts, because a reviewer who has read the rule reads facts as evidence — Chapter 22's reframe: show that the record contains what the standard asks for.

Evidence comes in four kinds and proves four things: the note proves the facts (enclose only the record at issue — minimum necessary); the policy proves the standard the payer owes itself; the manual proves the national standard; the edit proves what the payer's machinery permits — and indicator 1 is a permission, not an authorization. When the record implies but does not state, construct the argument from documented facts — never construct the record. A post-denial addendum asserting the disputed element is visible, dated, and worse than no appeal at all.

Commercial appeals run on the contract: one or two internal levels, windows of roughly 60 to 180 days running from the determination date — the clock ticks while the denial sits in a queue. Then the routes divide: external review for medical judgment, contract dispute provisions for money. Self-funded plans change the stack: ERISA, not state law; the DOL claims-procedure regulation; at least 180 days for the member's appeal; rights that belong to the member and are borrowed by authorization; a federal complaint route. Ask whether the plan is self-funded and write it down.

Medicare's five levels: redetermination (MAC, 120 days), reconsideration (QIC, 180 days), ALJ at OMHA (60 days, amount in controversy), the Medicare Appeals Council (60 days), judicial review (60 days, higher threshold). Rede- before recon-, MAC before QIC, independence starts at level 2, and the dollar thresholds start at level 3 and adjust annually. Levels 1 and 2 are the working practice's whole game — and the record substantially closes at reconsideration: all evidence in by level 2, or good cause later. Front-load everything.

The peer-to-peer is the treating physician's call, prepared by the billing office — the decision, the standard, the facts that meet it, the gap named honestly, the logistics — documented immediately, reference number and all. It lives mostly in the authorization and concurrent world; a post-service documentation denial is won in writing.

External review is the only reviewer the payer cannot overrule: an IRO, for medical-judgment denials, after internal exhaustion, within four months, binding on the plan — state process for insured plans, federal process for self-funded ones, and not the venue for pricing or edit disputes. Medicare's ladder has independence built in; Medicare Advantage auto-forwards upheld plan appeals to an independent review entity.

Track two calendars or lose to one of them. The appeal log carries the filing deadline, the payer's response deadline, and the next follow-up — opened at the decision to appeal, worked by date, followed up on a schedule, and closed only when the money posts. Filed is activity; decided, posted, and on time is performance. Outcomes flow back to Chapter 29's log as the overturn rate by category that decides next quarter's appeals.

And Account 10-4471's appeal won — submitted day 24, upheld day 59, paid day 66: \$98.40 plan, \$30.00 patient, \$128.40 whole. Three documents, four elements, one constructed paragraph, 31 minutes, and not one word created for the occasion. Q1 is confirmed by the payer's own reviewer. Q4 — whether any of this should have been necessary, and what it cost — is the last question standing, and it belongs to Chapter 40.

Key Terms

Appeal · timely filing for appeals · appeal letter structure · redetermination · reconsideration · levels of Medicare appeal · qualified independent contractor (QIC) · administrative law judge (ALJ) · Medicare Appeals Council · judicial review · amount in controversy · external review · independent review organization (IRO) · appeal log · authorized representative · escalation

Spaced Review

From Chapter 22 §22.6 and §22.11 — the reframe this chapter's letters are built on: stop arguing the service was needed; show that the record contains what the policy asks for. And Account 10-5502: why did three appeal letters fail where a phone call and four documented weeks succeeded?

From Chapter 22 §22.10 — the peer-to-peer review: who conducts it, and what is the coder's narrow role? This chapter added: who makes the call, and what goes on the one-page prep sheet?

From Chapter 29 §29.4 and §29.5 — run the ninety-second triage on a CO-50, and say why the queue sorts by deadline. This chapter's addition: which deadline, running from which date?

From Chapter 29 §29.6 and §29.10 — correcting versus appealing: what can a corrected claim cost you at some payers? And reconstruct the Encounter's three touches — 14 + 31 + 13 — and what each one produced.

From this chapter — the five levels in order, with decider and window; the evidence rule at reconsideration; the four kinds of evidence and what each proves; and the difference between constructing an argument and constructing a record.

Coming up: Chapter 31 — the \$185.00 line sat in accounts receivable for 49 days while all of this happened. Where it sat, what the aging report said about it, how the work queue surfaced it — and the arithmetic of the balances that are not worth what the Encounter's was.