Appendix I — The Certification Exam Study Guide

This is the book, rearranged. Forty chapters taught the work in the order money moves through it: the system, the record, the two great code sets, the edits and the prices, the claim, the refusal, the settings that play by different rules, and the audit. That order is right for learning. It is not the order you revise in, because a certification exam is organized by domain, and a candidate with six weeks left does not need a narrative — they need to know what is tested, which distinctions cost the most points, exactly where in this book each one lives, and whether they can still answer it without looking.

So this appendix does four things for each domain, and nothing else:

What is tested — the shape of the material, not a syllabus.
The distinctions candidates miss — harvested from the fifty-seven 🎓 Exam Watch callouts the book has been depositing since Chapter 1. Every one of them is somebody's actual wrong answer.
Review — the chapters and sections to open, by number.
Self-check — a short set of questions with compressed answers, to be attempted before you reopen anything.

Chapter 39 owns everything else, and this guide does not repeat it. Chapter 39 §39.3 and §39.4 map the credentials; §39.5 tells you how to choose between them on local evidence rather than on advice; §39.6 explains what the exams are built to measure; §39.7 gives the week-by-week preparation plan and the error log that makes it work; §39.8 handles the code books you may carry and what you may write in them; §39.9 covers accommodations, the clock, and the three-pass method; §39.10 the apprentice designation; §39.11 continuing education. If you are looking for the plan, the mechanics, or the choice of credential, you are in the wrong file. This one is the content review.

I.1 How to use this guide — and the one rule that governs every page of it

The rule, stated first because it outranks everything else here

Every parameter of your exam is set by the organization that administers it, and every one of them is revised. That includes — and this list is deliberately a list of kinds rather than of values — the number of items, the time allowed, the item formats, the way the exam is delivered, the materials you are permitted to have, the fee, the eligibility requirements, the retake and reschedule policy, the scoring method and the standard for passing, the domains covered, and the weight each domain carries.

This appendix states none of them, and you should distrust any study guide that does. A candidate who plans against a number that changed between one edition and their test date has been failed by the book, not by the exam. Chapter 39 §39.6 says the same thing about the exam and §39.8 says it about your code books, and the reason it is repeated a third time here is that this is the file a candidate reads in week eleven.

What to do instead: get the current candidate handbook and the current exam content outline for your specific exam, from the organization's own site, and read both end to end. The handbook governs every parameter. The content outline is your syllabus — it lists the domains and, usually, their approximate weights, and it is free. Then use this appendix beside it.

How the two documents fit together

The domains below are organized the way an exam is organized, but they are this book's domains, not any organization's. Your content outline may split, merge, name, or weight them differently. That is expected and it is not a problem, because the mapping runs both ways: find the topic on your outline, find it in the map at I.2, open the sections named there.

Weight your hours by your outline, not by this appendix's section lengths. I.7 is the longest section here because CPT is the largest body of material in the book; that is a fact about the book, not a prediction about your exam.

The revision loop that actually works

The failure mode in week nine is re-reading. Re-reading feels like studying and produces almost nothing, because recognizing material is not the same skill as retrieving it under a clock.

   THE LOOP — one domain, roughly 45 minutes

   1  READ the domain's "what is tested" paragraph.        2 min
      Nothing else. Close the book.

   2  ATTEMPT the self-check questions cold.               8 min
      Write the answers down. Writing them is the point --
      an answer you "sort of know" collapses when written.

   3  SCORE yourself against the compressed answers.       3 min
      Mark each one: KNEW IT / ALMOST / DID NOT.

   4  READ the "distinctions candidates miss" block.       5 min
      This is where your ALMOSTs usually are.

   5  OPEN only the sections behind your DID NOTs.        20 min
      Not the whole chapter. The section.

   6  LOG every miss in your error log, with the column   5 min
      Chapter 39 §39.7 adds: WHY you missed it.

Step 6 is the one people skip and it is the one that compounds. After forty entries the log stops being a list of facts and starts being a diagnosis — misread the stem, missed a guideline, wrong main term, did not verify in the Tabular, guessed under time pressure, genuinely did not know. Those have five different fixes and only one of them is "study more content."

What this guide is not

It is not a question bank. The self-checks are diagnostic, not representative; they exist to find your gaps, not to simulate an exam. Full-length timed simulations are §39.7's week eleven and nothing here substitutes for them.

It is not a code list, and you may not assign from it. Every code named below is named for its teaching purpose — a contrast, a trap, a family boundary. ICD-10-CM changes every October 1, CPT every January 1, HCPCS Level II quarterly, and the National Correct Coding Initiative (NCCI) edits quarterly. Appendix A collects the structure of all three; the current book or encoder is the only thing you assign from, on the exam and at the desk.

And it is not a substitute for having done the work. Chapter 39 §39.6 puts it plainly: the exams test whether you can find and apply the answer, not whether you remember it. A guide can tell you where the answer lives. Only repetitions make you fast enough to get there.

I.2 The domain map — this book, rearranged

   DOMAIN                                    CHAPTERS            HERE
   ---------------------------------------   -----------------   ----
   1  The revenue cycle and reimbursement     1, 2, 3            I.3
   2  Compliance and the regulatory frame     5, 3, 32           I.4
   3  The medical record and documentation    4, 38              I.5
   4  ICD-10-CM                                                  I.6
        structure                             7
        the lookup discipline                 8
        the Official Guidelines               9
        body-system applications              10, 11
        the special chapters                  12
   5  CPT                                                        I.7
        structure                             13
        modifiers                             14
        evaluation and management             15, 16
        surgery                               17, 18
        radiology, path and lab, medicine     19
   6  HCPCS Level II                          20                 I.8
   7  Edits, medical necessity, pricing       21, 22, 23         I.9
   8  Claim forms and electronic submission   24, 25, 26, 27     I.10
   9  Remittance, denials, appeals, AR        28, 29, 30, 31, 32 I.11
  10  Facility coding                         33, 34             I.12
  11  Specialty and risk-adjusted coding      35, 36             I.13
  12  Auditing and documentation integrity    37, 38             I.14

   PLUS   The high-yield distinctions ....................... I.15
          The navigation drills ............................. I.16

   READ ACROSS, NOT DOWN. Domains 4 and 5 are the largest blocks on
   most coding exams and the smallest on a billing-oriented one. Your
   own content outline decides; this column of chapter numbers does not.

Two structural notes before the domains.

The professional/facility line runs through everything. Domains 4 through 7 are mostly written from the professional claim's side, because that is where most candidates start; Domain 10 is the facility counterpart, and Chapter 16 §16.9 is the section that shows one encounter producing both. A facility-oriented exam does not merely add Domain 10 — it re-weights Domains 4 and 5 around it.

And billing is not a subset of coding. Domains 8, 9, and parts of 1 are the whole content of a billing credential and a light touch on a coding one. If your outline calls them a third of the exam, they are a third of your hours, no matter how much more interesting the surgery section is.

I.3 Domain 1 — The revenue cycle and reimbursement

What is tested. Vocabulary with money attached to it, and the order in which things happen. The four numbers a claim produces and how they relate. The structure of insurance benefits. Plan types and what each one requires of a patient before a service. Who is primary when two plans cover one person. The government programs and which part of Medicare pays for what. This domain is where a candidate loses points not to difficulty but to imprecision — two words that mean different things get used interchangeably and the distractor is waiting.

The distinctions candidates miss.

Charge, allowed amount, payment, patient responsibility are four different numbers and never synonyms. Charge is what the provider bills, allowed is what the contract says the service is worth, payment is what the payer sends, patient responsibility is the part of the allowed amount the patient owes. Chapter 1 §1.2.
Payers pay the lesser of the billed amount and the allowed amount. A stem giving a charge of \$90.00 against an allowed amount of \$96.52 is testing exactly this, and the answer is \$90.00 — Chapter 23 §23.9's most-tested fact.
Self-funded is a funding arrangement; capitation is a payment arrangement. The classic stem describes an employer paying claims from its own funds with an insurer doing only administration, and offers "capitated" as the distractor. Keep the axes apart: who bears the risk of the population (Chapter 2 §2.5) versus how the provider is paid (Chapter 2 §2.6).
The birthday rule reads month and day only. The year is in the stem to make you choose the older parent, and choosing the older parent is wrong. It applies to dependent children of parents who are married or living together; if the stem says divorced, a court decree controls where one exists. Chapter 2 §2.8.
Medicare Part B covers outpatient hospital services. "Hospital equals Part A" is the reasoning that produces the wrong answer. The rule is not building-based: an inpatient admission is Part A; an outpatient encounter at the same hospital is Part B, billed on a UB-04 and paid under the Outpatient Prospective Payment System. Chapter 3 §3.2.
The Part A deductible is per benefit period, not per year, and a benefit period ends after 60 consecutive days out of an inpatient setting. Expect a stem with two admissions and a gap, asking how many deductibles are owed. Chapter 3 §3.2.
Original Medicare has no out-of-pocket maximum. A distractor will offer a figure. Chapter 3 §3.2.
Medicare Secondary Payer turns on active employment, not on merely having coverage. Twenty or more employees plus current employment status makes the group health plan primary on the age-65 basis; retiree coverage is not current employment status. The 100-employee threshold belongs to the disability entitlement basis. Keep each threshold attached to its category. Chapter 3 §3.8.
A Medicare Administrative Contractor applies national policy and fills gaps locally. It does not set national policy and cannot override a national coverage determination. Chapter 3 §3.3.

Review. Chapter 1 §1.2, §1.3, §1.5 · Chapter 2 §2.2, §2.3, §2.4, §2.5, §2.6, §2.8 · Chapter 3 §3.2, §3.3, §3.4, §3.6, §3.7, §3.8 · Chapter 23 §23.9 for the lesser-of rule.

Self-check.

A service is charged at \$185.00, the contract allows \$128.40, the patient's copayment is \$30.00. Name all four numbers and state the plan's payment.
A patient is 68, still working for a company with 42 employees, and covered by its group plan. Which payer is primary? Change the company to 12 employees — now which?
What is the difference between a fee schedule and a percentage-of-billed-charges arrangement, and which one is the only case where raising the charge raises the payment?
A child is covered by both parents' plans. The mother's date of birth is in August, the father's in April, and the father is older. Which plan is primary and why is the age irrelevant?
A Medicare beneficiary is seen in a hospital outpatient clinic. Which part of Medicare pays, and on which claim form does the hospital bill?

Answers: 1 — charge \$185.00, allowed \$128.40, patient responsibility \$30.00, plan payment \$98.40; the \$56.60 gap is the contractual adjustment and nobody owes it. 2 — the group health plan, because the employer has 20 or more employees and the beneficiary is actively working; at 12 employees it flips to Medicare. 3 — a fee schedule sets an amount per code independent of the charge; percentage of billed charges is the one arrangement where the charge drives the payment, which is why it is increasingly rare (Chapter 2 §2.6, Chapter 23 §23.7). 4 — the father's, because April precedes August; the rule reads month and day. 5 — Part B, on the UB-04.

I.4 Domain 2 — Compliance and the regulatory frame

What is tested. Named statutes and what each one prohibits. The difference between an error and fraud. The seven elements of a compliance program. HIPAA as three separate things. And the newer patient-facing federal rules, which have arrived on billing-oriented exams.

The distinctions candidates miss.

Fraud requires intent; abuse does not. The line between an error and fraud is intent and pattern, and the line between an error and a reportable overpayment is sixty days. Chapter 5 §5.2, Chapter 31 §31.9.
The seven elements are tested by exclusion. Expect "which of the following is NOT one of the seven elements," with distractors like obtaining professional liability insurance or conducting employee satisfaction surveys. And expect the pairing question: auditing and monitoring detects; prompt response and corrective action responds. They are separate elements for that reason. Chapter 5 §5.6.
HIPAA is three things — Privacy, Security, and the Transactions and Code Sets standards — and a question about the 837 is a transactions question, not a privacy one. Chapter 5 §5.7, Chapter 27 §27.1.
Billing and coding are "payment," not "treatment." That is why you may access a record without the patient's authorization, and it is also why the minimum necessary standard applies to you: it does not apply to treatment disclosures, and it very much applies to payment and operations. Chapter 5 §5.7.
Upcoding and downcoding are both errors. Downcoding is not the conservative option, it is not a defense, and it is not safe. Chapter 5 §5.8, and see I.15.
Cost sharing lives inside the allowed amount; a balance bill lives above it. A stem that calls a deductible a balance bill is testing exactly that, and collecting a protected patient's in-network coinsurance is legal and correct. Chapter 32 §32.4.
Some out-of-network services can never be consented into balance billing. The ancillary categories — anesthesiology, pathology, radiology and their companions — are a favorite stem because they are the counterintuitive half of the rule. Chapter 32 §32.4.
Independent dispute resolution is between the plan and the provider. A stem with the patient participating in it has described the wrong process; the patient's analogue is the patient-provider dispute process attached to good faith estimates. Chapter 32 §32.3, §32.4.

Review. Chapter 5 §5.2, §5.3, §5.4, §5.5, §5.6, §5.7, §5.8 · Chapter 3 §3.4 for assignment and the limiting charge · Chapter 32 §32.3, §32.4, §32.5 · Chapter 31 §31.9 for the sixty-day rule · Appendix F for the statutes with their citations.

Self-check.

Two claims are submitted for services that were not documented. In one case the coder misread the note; in the other, a manager instructed the coder to bill that way regardless of the note. Which statutes are in play, and what makes the second one different?
Which element of a compliance program covers responding to a problem you have already found?
A practice discovers it has been overpaid on a pattern of claims. What clock starts, and how long is it?
Is collecting a \$40.00 coinsurance from a patient protected against surprise billing lawful?
Name the three HIPAA rules and say which one governs the 835.

Answers: 1 — both engage the False Claims Act's theory of a false claim, but intent and pattern are what separate error from fraud (Chapter 5 §5.2, §5.3); the second also engages the coder's own professional obligation and the escalation path in Chapter 5 §5.9. 2 — element seven, prompt response and corrective action. 3 — the sixty-day overpayment clock (Chapter 31 §31.9). 4 — yes; cost sharing sits inside the allowed amount and is unaffected. 5 — Privacy, Security, and Transactions and Code Sets; the 835 is a transactions-standard question.

I.5 Domain 3 — The medical record and documentation

What is tested. What each document type contains and which one answers a given question. What makes a record legally sufficient. And — the single most-tested idea in this domain — what a coder may and may not conclude from a note.

The distinctions candidates miss.

The shape of the item is almost always the same: a scenario describes a service, the documentation is subtly insufficient for the obvious code, and the correct answer is the less specific code or a query. Chapter 4 §4.7.
The reliable tell is a finding that appears only in the history or only in the patient's report, never in the provider's assessment. "The patient states they were told they have osteoarthritis" is not a diagnosis of osteoarthritis. Chapter 4 §4.7.
The second reliable tell is laterality. If a scenario carefully avoids saying which side, the unspecified code is the answer — and on a real claim, a query. This is why the book contrasts M25.561, M25.562, and M25.569 as a set. Chapter 4 §4.7, Chapter 7 §7.7.
You may climb the specificity ladder exactly as far as the documentation carries you. Going higher is inference and is not permitted; staying lower is downcoding and is also an error. Chapter 4 §4.7, Chapter 5 §5.8.
The clinician owns the clinical statement. A stem in which a coder adds, moves, or infers a diagnosis is always wrong, however obvious the clinical picture. Chapter 38 §38.9.
A query is judged on whether it supplies the answer, not on whether it is polite. A query can be courteous, well-formatted, clinically sensible, and still leading, because it names a code, names a payment consequence, stacks the options, or offers no way to decline. Read the option list last and read it twice. Chapter 38 §38.3, §38.9.
Do not query to obtain a higher-paying code, when the answer is already in the record, or when the honest resolution is the less specific code. Chapter 4 §4.9, Chapter 38 §38.9.
Concurrent review means during the encounter, not shortly after it. Chapter 38 §38.2, §38.9.

Review. Chapter 4 §4.2, §4.4, §4.5, §4.6, §4.7, §4.9 · Chapter 38 §38.2, §38.3 · Appendix G for the query templates.

Self-check.

An operative report's header says one procedure and the body of the report documents a different objective. Which one do you code?
A note's history section mentions a chronic condition; the assessment does not address it. May it be reported on this encounter's claim?
Name three things that make a query leading.
What must a signature do before the record it signs will support a claim?
A record documents a fracture but not which side. What do you assign, and what do you do at the desk?

Answers: 1 — the objective documented; the label is not the record (Chapter 33 §33.9 makes the same point on the facility side). 2 — no; Chapter 9 §9.7 and Chapter 36 §36.7 both turn on whether the condition was addressed at the encounter. 3 — supplying the answer, naming the code or the payment consequence, offering options with no way to decline (Chapter 38 §38.3). 4 — identify the author and authenticate the entry, legibly and datably (Chapter 4 §4.4). 5 — the unspecified code, and a query.

I.6 Domain 4 — ICD-10-CM

This is the largest single block on most diagnosis-bearing exams and it is where routing beats recall most decisively. Five sub-domains.

The structure of a code

What is tested. Character positions and what each one carries; which letters live in which chapter; the placeholder; the seventh character; laterality; combination codes.

The distinctions candidates miss.

Count the characters, ignoring the decimal point. A code carrying a seventh character must have exactly seven. The placeholder question comes in two forms — build the code (W19A / W19.XA / W19.XXA / W19.XXXA, and only one of those is right) or identify the invalid code among four. Chapter 7 §7.5.
H is the only letter covering two chapters — the eye and the ear — which is precisely why the letter-range question uses it. Chapter 7 §7.2.
Chapter priority is a rule of thumb; the Tabular is the authority. The general answer is that the special chapters take precedence over the body-system chapters, but the real answer is that Excludes and instructional notes tell you, and the exam rewards verify in the Tabular over reciting a hierarchy. Chapter 7 §7.2.
"Unspecified" is always available and always defensible when the documentation stops there. Chapter 7 §7.9.

Review. Chapter 7 §7.2, §7.3, §7.4, §7.5, §7.6, §7.7, §7.8, §7.9 · Appendix A.

The lookup discipline

What is tested. The two-step rule, and the conventions that live in each of the two books.

The distinctions candidates miss.

Parentheses are permission to ignore; indentation is a requirement to document. A stem documenting "benign essential hypertension" and offering three more specific-looking codes is testing nonessential modifiers, and the answer is I10. The mirror stem documents "pain" with no site and offers a joint-specific code; you may not descend to a subterm the documentation does not support. Chapter 8 §8.3.
Excludes2 is permission, not prohibition. It says the code you are on does not cover the other thing — so if the patient has the other thing, you need a second code. Excludes1 says the two cannot be reported together. Both a definitional item and an applied item are likely. Chapter 8 §8.5.
"Code first" and "use additional code" are two ends of one relationship and both fix the sequence. "Code also" requires both codes and deliberately declines to set the order, because that depends on the reason for the encounter. Chapter 8 §8.6, Chapter 9 §9.9.
NEC blames the book; NOS blames the note. The condition is documented specifically but the classification has no code for it — NEC. The documentation does not specify — NOS. Chapter 8 §8.7.
Brackets mean two different things in the two books: explanatory text in the Tabular, and a manifestation code that must be sequenced second in the Index. Chapter 8 §8.8.
"And" means and/or. "With" means associated with or due to — and can create an assumed causal relationship. "See" is mandatory; "see also" is advisory. Chapter 8 §8.4, §8.8.

Review. Chapter 8 §8.1 through §8.10, all of it. This is the highest-yield chapter in the book per hour spent.

The Official Guidelines

What is tested. The four sections and what each governs; first-listed versus principal; the uncertain-diagnosis rule; signs and symptoms; the "with" convention; sequencing.

The distinctions candidates miss.

Read the setting before you read the clinical facts. Office, clinic, emergency department, and ambulatory surgery center are all outpatient, governed by Section IV. The trap that catches the most people is that the emergency department is outpatient — a patient seen there and sent home is an outpatient encounter regardless of how sick they were. Chapter 9 §9.5.
The uncertain-diagnosis rule reverses between settings. "Probable," "suspected," "likely," "rule out" are coded as though established on an inpatient record and are not coded at all in the office. And the inpatient half says documented at the time of discharge — an uncertain diagnosis that appears in a day-two progress note and is abandoned by discharge is not the same thing. Chapter 9 §9.5.
Signs and symptoms integral to the disease are not coded separately; signs and symptoms not routinely associated with it are. People remember only the first half. Chapter 9 §9.6.
The "with" convention is a question about the classification, not about the clinic. The discriminating question is always has the classification linked these, in the Index or in a Tabular instruction? If yes, presume. If no, you need documentation. Candidates keep trying to answer it clinically. Chapter 9 §9.7.
Acute and chronic, both documented, both with subentries at the same indentation level: code both, acute first. Chapter 9 §9.8.

Review. Chapter 9 §9.2, §9.3, §9.4, §9.5, §9.6, §9.7, §9.8, §9.9 — and the current Guidelines themselves, which are republished every year and are free.

Body-system applications

What is tested. Applied coding across the systems, with the sequencing rules that ride on top.

The distinctions candidates miss.

Sepsis is not severe sepsis, and the difference is documented acute organ dysfunction. The underlying systemic infection sequences first. A stem documenting sepsis and acute kidney injury is testing whether you assume the association: severe sepsis requires the dysfunction to be documented as associated with the sepsis. Chapter 10 §10.3.
The neoplasm behavior column comes from the pathology report or the provider's statement. You must know the column before you can use the Table. A coder who guesses the column has guessed the diagnosis. Chapter 10 §10.4.
An encounter solely for chemotherapy sequences the encounter code first, with the malignancy second — the distractor is always the cancer, because it feels more important. The rule is about the reason for the encounter. And note the variant: if the patient is also treated for something else, the encounter is no longer solely for chemotherapy. Chapter 10 §10.4, Chapter 9 §9.3.
A history code requires three things at once: excised or eradicated, no further treatment directed to the site, and no evidence remaining. A patient on adjuvant therapy does not have a history code. Chapter 10 §10.5.
The hypertension asymmetry is the single most reliable item in circulatory coding. Hypertension plus a heart condition with no stated relationship: two codes. Hypertension plus chronic kidney disease with no stated relationship: the combination code, because the relationship is presumed. Each stem's distractor is the other rule's answer. Chapter 11 §11.1.
Obstetric coding clusters on four things: the obstetric code sequences first unless the provider documents the pregnancy as incidental; the trimester comes from documented weeks of gestation at the encounter; the seventh character here identifies the fetus, not the episode of care; and the outcome-of-delivery code goes on the maternal record only. Chapter 11 §11.9.

Review. Chapter 10 §10.3, §10.4, §10.5, §10.6 · Chapter 11 §11.1, §11.2, §11.5, §11.8, §11.9 — plus the worked chart sets at Chapter 10 §10.10 and Chapter 11 §11.10, which are the closest thing in the book to timed practice on this material.

The special chapters

What is tested. Injuries and their seventh characters, burns, the poisoning table, external causes, symptom codes, and Z-codes.

The distinctions candidates miss.

"Initial" does not mean first visit and "subsequent" does not mean second visit. The distinction is active treatment versus routine care during healing. A patient seen by a new provider for the first time, two weeks after a fracture was set elsewhere and healing normally, is D — and the distractor is A, because it is the first visit to that provider. A patient returning for the third time during active treatment of a complex wound is still A. Chapter 12 §12.3, Chapter 7 §7.6.
Sequela takes two codes, the residual condition first and the injury with S second — a scar contracture following a burn is the standard stem. Chapter 12 §12.3.
Nothing rejects when you get this wrong, which is why it is worth exam points and audit exposure at the same time: both characters are valid codes and the claim pays, while the record now reads as a series of new acute events. Chapter 12 §12.3.
One table, four columns: poisoning, adverse effect, underdosing, toxic effect. The same drug appears in more than one column and the column is chosen by what happened, not by the drug. Chapter 12 §12.6.
Z-codes are not a fallback. Screening, status, history, and aftercare each have their own sequencing consequences, and screening versus diagnostic is the distinction that decides both a code and, later, a patient's bill. Chapter 12 §12.9, Chapter 34 §34.11.

Review. Chapter 12 §12.2, §12.3, §12.4, §12.6, §12.7, §12.9, §12.10.

Self-check for Domain 4.

A stem documents a fall at home with no further detail, initial encounter. Build the code's character positions and say how many placeholders you need.
The Tabular shows an Excludes2 note under the code you have reached, naming a condition the patient also has. How many codes do you report?
A patient in the office has "probable pneumonia" documented. A second patient's discharge summary says the same thing. Code each.
Hypertension and chronic systolic heart failure, no relationship stated. Hypertension and stage 4 chronic kidney disease, no relationship stated. Both, please.
An orthopedist sees a patient for the first time two weeks after an emergency department set a fracture; healing is normal. Which seventh character?
A patient presents solely for chemotherapy for a known malignancy. What is sequenced first?

Answers: 1 — seven positions, the seventh character carries the episode of care, so you need placeholders through the sixth. 2 — two; Excludes2 is permission. 3 — office: code the documented signs or symptoms, not the probable diagnosis; discharge summary: code it as though established. 4 — heart: two separate codes; kidney: the combination code plus the stage code. 5 — D, subsequent encounter; active treatment is complete. 6 — the encounter code for the chemotherapy, malignancy second.

I.7 Domain 5 — CPT

The largest block on a professional-coding exam. Five sub-domains, and the surgery block is usually the heaviest of them.

The structure of the book

What is tested. The categories, the sections and their boundaries, the symbols, the parentheticals, add-on codes, "separate procedure," and unlisted codes.

The distinctions candidates miss.

Format identification is free points. Five digits — Category I. Ending in F — Category II. Ending in T — Category III. And the mandatory-use rule: if a Category III code describes the procedure, report it rather than an unlisted Category I code. The distractor is always the unlisted code, because it feels more official. Chapter 13 §13.2.
All add-on codes are exempt from modifier 51, but not all modifier-51-exempt codes are add-on codes. Two different lists in two different appendices of the code book, and candidates conflate them. Chapter 13 §13.5, §13.7.
The book is not in numeric order. Evaluation and management sits at the front and carries the highest numbers. Chapter 13 §13.3.
"Separate procedure" means reportable only when performed independently or when unrelated to or distinct from other procedures at the same session. The distractor is "always, because it is separate," which is the reading the name invites. Chapter 13 §13.8.
Parentheticals are binding, including the ones that prohibit reporting two codes together — a bundling instruction living inside the code book itself, independent of any edit file. Chapter 13 §13.6.
An unlisted code requires a special report and does not price itself, which is why the report is what the payer uses to decide payment. Chapter 13 §13.9.

Review. Chapter 13 §13.2, §13.3, §13.4, §13.5, §13.6, §13.7, §13.8, §13.9.

Modifiers

What is tested. What each common modifier asserts, which one applies to a described circumstance, and the sequencing of several on one line. This is the sub-domain where a candidate can gain the most points fastest, because the rules are finite.

The distinctions candidates miss.

Modifier 25 does not require a different diagnosis. The distractor always suggests it does. What it requires is that the evaluation and management service be significant and separately identifiable — above and beyond the usual pre- and post-operative work of the procedure. If the scenario describes only the evaluation that led to the minor procedure, the answer is no modifier 25 and no separately reported service. Chapter 14 §14.4.
25 is for minor procedures (000 and 010 global periods); 57 is for the decision for major surgery (090). Candidates swap them, and the global period in the stem is what decides. Chapter 14 §14.4, §14.9, Chapter 17 §17.2.
51 and 59 are not alternatives. 51 says multiple procedures were performed; 59 says this service was distinct. One is about how many; the other is about whether. Chapter 14 §14.6.
59 is a last resort. If a more descriptive modifier is available — an anatomic modifier, a global-period modifier, one of the X{EPSU} subset — use that instead. Chapter 14 §14.5.
26 and TC split one service between two entities. A radiologist reading a study performed at a hospital reports the interpretation with modifier 26; the hospital reports the technical component. Nobody reports the global service, because no single entity furnished all of it, and a candidate who reports the global code has billed for somebody else's equipment. Chapter 14 §14.7, Chapter 19 §19.1.
54, 55, and 56 turn on who did what. Operated and transferred the follow-up: 54. Assumed the follow-up without operating: 55. Preoperative management only: 56. Both physicians report the same procedure code — there is no evaluation and management code for the one providing postoperative care, because that care is inside the global package. And a transfer of care must be agreed and dated; a physician who simply stops seeing the patient has not created a 54/55 split. Chapter 18 §18.8a.
Reporting modifier 50 on a code that is already inherently bilateral is a duplicate, and reporting a unilateral code once for a bilateral procedure is an underpayment. The descriptor decides. Chapter 18 §18.9, Chapter 14 §14.8.

Review. Chapter 14, all twelve sections — and Appendix B, which is the modifier reference this domain is built on.

Evaluation and management

What is tested. Level selection under the current framework; new versus established; time; the setting-specific families.

The distinctions candidates miss.

99201 no longer exists. It was deleted effective January 1, 2021. The new-patient family therefore has four levels and the established family five, and that asymmetry is a favorite item that looks like a typographical error until you know why. If a study resource lists 99201, be suspicious of everything else it says about evaluation and management. Chapter 15 §15.2, §15.3.
99211 has no medical decision making requirement and no time requirement, and may not require the presence of a physician or other qualified health care professional. It is the only office code like that; 99281 is the emergency department family's analogue. Chapter 15 §15.2, Chapter 16 §16.6.
Time counts only the reporting professional's time, only on the date of the encounter, and excludes time spent on a separately reported service. Three stems recur: the nurse's twenty minutes plus the physician's fifteen is fifteen; twenty-five minutes today plus ten tomorrow is twenty-five; forty minutes of which twelve were a separately reported procedure is twenty-eight. Chapter 15 §15.8.
Time and medical decision making are alternatives, and time never caps a decision-making level. A note supporting 99214 by decision making and 28 minutes by time is a 99214. Downcoding it because "the time says 99213" misunderstands the rule in a way that costs an organization money across a very large number of claims. Chapter 15 §15.13.
The three Rs make a consultation: a documented request, a documented reason, and a written report back. Read the stem for the missing one. A self-referred patient is not a consultation; a requesting physician who transferred management of the problem has made it a transfer of care; a consultant who initiates treatment is still a consultant if the three Rs are met. And the payer question: where consultation codes are not recognized, report the setting-appropriate service instead. Chapter 16 §16.5.
The emergency department family has no new/established distinction and no time-based option. Leveling is by medical decision making only. Chapter 16 §16.6.
Observation is an outpatient service, and the two-midnight benchmark turns on the physician's documented expectation, not on hindsight. Condition code 44 changes status only before discharge. Chapter 16 §16.3, Chapter 34 §34.10.
"Initial" hospital care means the first service by that professional or a same-specialty group partner, not the admission. Chapter 16 §16.2.
The prolonged-services complication is real: CPT provides 99417, attached only to the highest-level codes and in full increments; Medicare created G2212 with a different starting threshold, so fewer encounters qualify. Chapter 15 §15.9.

Review. Chapter 15 §15.2, §15.4, §15.5, §15.6, §15.7, §15.8, §15.9, §15.10 · Chapter 16 §16.2, §16.3, §16.5, §16.6, §16.7.

Surgery

What is tested. The surgical package and global periods; the integumentary measurement and repair rules; musculoskeletal and fracture care; endoscopic families; multiple-procedure order; anesthesia.

The distinctions candidates miss.

The measurement rule: lesion diameter plus the narrowest margins, measured before excision. The stem will contain all four numbers with three of them as distractors. Say the rule out loud before you look at the answers. Multiple lesions are coded separately, each with its own measurement — they are not added together. And excision includes simple closure, so reporting a simple repair alongside it is unbundling. Chapter 17 §17.4, §17.5.
Shave removal, destruction, and excision are different families. Destruction produces no specimen, therefore no pathology report, therefore the family has to make the benign/malignant distinction on clinical grounds. Chapter 17 §17.4.
A biopsy of a lesion that is then excised at the same session is generally not separately reported. Chapter 17 §17.4.
The cardiovascular section is not tested on expertise; it is tested on bundling. A catheterization scenario listing the components separately is asking whether you will report them separately. You should not. The same trap appears with vascular interventions where the intervention code includes the catheter placement and the angiography. Chapter 18 §18.4.
Guidance has a three-step decision. Is guidance named in the descriptor? Included. Does a separate code exist for the procedure with guidance? Use that code. Otherwise, is there a standalone guidance code whose own requirements are met? This is exactly why 20610 and 20611 are the book's standing contrast, and why every guidance code requires permanent image recording and a report. Chapter 19 §19.3, Chapter 17 §17.7.
The order of lines matters. The highest-valued procedure goes on the first line, because multiple procedure reductions apply to the lines beneath it. Chapter 18 §18.8, Chapter 25 §25.4.
Anesthesia uses base units plus time units, and the stem will give you both room time and anesthesia time and hope you use the wrong one. Chapter 18 §18.11, Chapter 35 §35.7.

Review. Chapter 17 §17.1, §17.2, §17.4, §17.5, §17.7 · Chapter 18 §18.1, §18.2, §18.4, §18.8, §18.8a, §18.9, §18.11.

Radiology, pathology and laboratory, and medicine

What is tested. The component split; views and contrast; the panel rule; repeat testing; the administration hierarchies.

The distinctions candidates miss.

The contrast rule, and the distractor is always oral contrast. "With contrast" means contrast administered by the routes the section's guidelines define; a study performed with oral contrast only is not coded as "with contrast" under those rules. Chapter 19 §19.2.
Count the views. A scenario stating three views of the knee is telling you the code, and a candidate who skims past it picks the two-view code. Chapter 19 §19.2.
Some radiology codes are inherently bilateral and some are not — the same descriptor discipline as Chapter 18 §18.9. Chapter 19 §19.2.
A panel is one code. If every component of a panel was performed, report the panel, not the components; reporting the components separately is unbundling. And a unique test for decision-making purposes is a CPT code, so a panel counts as one test. Chapter 19 §19.4, Chapter 15 §15.6.
Modifier 91 is for a repeat test to obtain subsequent results, not for a repeat because the first one failed. Chapter 19 §19.7.
Immunizations take two codes — the vaccine product and the administration — and the vaccine scenario missing its administration line is a standard stem. Chapter 19 §19.10, Chapter 35 §35.5.

Review. Chapter 19 §19.1, §19.2, §19.3, §19.4, §19.6, §19.7, §19.9, §19.10, §19.11.

Self-check for Domain 5.

A minor procedure with a 000-day global period is performed at an office visit where three unrelated chronic conditions were also addressed. Which modifier, and on which line?
A radiologist interprets a study performed on hospital equipment. What does the radiologist report, what does the hospital report, and what does nobody report?
A physician documents 40 minutes of total time, 12 of which were spent performing a separately reported procedure. What time supports the level?
A lesion measuring 1.2 cm is excised with 0.3 cm margins on each side. What measurement selects the code?
A consultant is asked by an attending to evaluate a patient, does so, initiates treatment, and writes back. Consultation or not?
A code is designated a separate procedure and was performed as part of a larger procedure at the same site. Report it?

Answers: 1 — modifier 25 on the evaluation and management line, not on the procedure. 2 — the professional component with modifier 26; the technical component; nobody reports the global service. 3 — 28 minutes. 4 — 1.2 + 0.3 + 0.3 = 1.8 cm, measured before excision. 5 — yes, if the request, the reason, and the written report back are documented; initiating treatment does not disqualify it. 6 — no.

I.8 Domain 6 — HCPCS Level II

What is tested. What the code set covers, the letter families, and — far above everything else — units.

The distinctions candidates miss.

Read the descriptor for the dosage, divide the dose given, report whole units. That is the whole arithmetic and it is on every professional-coding exam in some form. Chapter 20 §20.3.
A "per 10 mg" descriptor and a 100 mg dose is ten units, not one. Candidates who recognize the drug name stop reading at the comma. Chapter 20 §20.3.
If a code exists for the exact dose administered, use it rather than multiplying a smaller one — which is why J1030 and J1040 are the book's standing pair.
The dose is not the vial. A 100 mg vial from which 60 mg was administered is 60 mg administered; the remaining 40 mg is a waste question (JW and JZ), not a units question. Chapter 20 §20.4.
HCPCS Level II changes quarterly — four times a year, against organizations that handle code updates once. A terminated code sitting in a charge file produces denials nobody traces. Chapter 20 §20.1.
Where CPT and HCPCS Level II both have a code, payer instruction decides, and Medicare frequently requires the G-code. Chapter 20 §20.9 sets the rule; G2212 is the book's worked example of it, and Chapter 15 §15.9 shows what the different threshold does to a prolonged-service claim.

Review. Chapter 20 §20.1, §20.2, §20.3, §20.4, §20.5, §20.6, §20.9, §20.10 · Appendix A for the letter families.

Self-check. A descriptor reads "per 10 mg." The physician administers 60 mg from a 100 mg single-dose vial and discards the remainder. How many units of the drug code, and what happens to the remainder? (Six units; the discarded 40 mg is a waste question governed by the JW and JZ rules and by payer policy — Chapter 20 §20.4.)

I.9 Domain 7 — Edits, medical necessity, and pricing

What is tested. Why two codes may not be reported together and what may override that; whether a service is covered at all; and how a code becomes a dollar figure.

The distinctions candidates miss.

The standards-of-practice items all look the same — a procedure plus something on the list, asking how many codes — and the answer is almost always one. Local anesthesia by the operating physician: included. The surgical approach: included. Wound irrigation and simple closure: included, unless the closure rises to a separately reportable repair. Chapter 21 §21.6, Chapter 17 §17.5.
Comprehensive/component and mutually exclusive are different rationales, and knowing which one you are looking at decides what argument is even available. A different session, site, lesion, or encounter is the argument that sometimes works against a comprehensive/component edit, and that is precisely what modifier 59 and the X{EPSU} subset assert. Chapter 21 §21.7, Chapter 14 §14.5.
The modifier indicator decides whether an override is possible at all, before any question of whether it is justified. Chapter 21 §21.3.
Medical necessity is a coverage concept, not a clinical one. See I.15; it is the most consequential single sentence in this domain. Chapter 22 §22.1.
The diagnosis codes that support a service live in the billing and coding article, not in the local coverage determination and not in the national one. Candidates reliably answer LCD. Chapter 22 §22.4.
When a national and a local determination conflict, the national governs, and an LCD generally does not exist where an NCD applies. Chapter 22 §22.3.
A statutorily excluded service does not need an Advance Beneficiary Notice, because the notice exists for services that would be covered but for the reasonable-and-necessary determination. GY, and GX if a voluntary notice was given. Chapter 22 §22.8, §22.9.
An Advance Beneficiary Notice signed at checkout, saying only that Medicare may not pay, is defective — not in advance, and no specific reason — and GA is not supportable on it. Chapter 22 §22.8.
"Once per calendar year" and "once every 12 months" are different rules, and the exam will give you dates chosen to separate them. Chapter 22 §22.7.
The relative value arithmetic is done in two steps, every time: adjust each of the three components by its own geographic index, sum, then multiply by the conversion factor. Candidates who sum the relative value units first and apply one index get a wrong answer that looks plausible. Chapter 23 §23.2, §23.3.
"Facility or non-facility" is a question about who supplied the overhead, and the answer selects the practice expense component. Chapter 23 §23.5.
Gross collection rate uses charges as the denominator; net uses what you were entitled to collect. Exams ask which is meaningful, and the answer is net — a practice that doubles its charges halves its gross rate without any change in performance. Chapter 23 §23.10.

   THE TWO-STEP RELATIVE VALUE CALCULATION
   [constructed teaching figures -- verify current values in the
    Medicare Physician Fee Schedule before you quote any of them]

     work 1.92 x GPCI 1.000 ....................  1.92000
     practice expense 0.89 x GPCI 1.008 ........  0.89712
     malpractice 0.13 x GPCI 1.005 .............  0.13065
                                                 --------
     adjusted total ............................  2.94777
     x conversion factor 32.7442 ...............  96.5224
                                                 --------
     allowed .................................... $96.52

   THE WRONG ANSWER THAT LOOKS RIGHT:
     (1.92 + 0.89 + 0.13) = 2.94, x one index, x CF.
     Close enough to be on the answer list. Not the method.

Review. Chapter 21 §21.2, §21.3, §21.4, §21.6, §21.7, §21.8 · Chapter 22 §22.1, §22.3, §22.4, §22.6, §22.7, §22.8, §22.9 · Chapter 23 §23.2, §23.3, §23.4, §23.5, §23.10.

Self-check.

Two codes hit a procedure-to-procedure edit with a modifier indicator of 0. What are your options?
Where do you find the diagnosis codes that support a service under a Medicare local policy?
A service is excluded by statute. Is an Advance Beneficiary Notice required, and which modifier applies?
A code has work 1.92, practice expense 0.89, malpractice 0.13. Describe the calculation in the right order.
Which collection rate can be improved by raising charges, and what does that tell you about it?

Answers: 1 — none involving a modifier; the edit cannot be overridden, so the question becomes whether the second service should have been reported at all (Chapter 21 §21.3). 2 — the billing and coding article. 3 — not required; GY, plus GX if a voluntary notice was given. 4 — adjust each component by its own geographic index, sum the three, then multiply by the conversion factor. 5 — neither, honestly; but the gross rate appears to worsen when charges rise, which is why it measures nothing about collection performance.

I.10 Domain 8 — The claim forms and electronic submission

What is tested. Which form goes with which kind of provider, what specific fields hold, the electronic transaction numbers, and where a claim can fail before anyone adjudicates it.

The distinctions candidates miss.

CMS-1500 for professional and non-institutional providers; UB-04 for institutional. The UB-04's official designation is CMS-1450 and exams use both names. The electronic equivalents are 837P and 837I. The CMS-1500 is maintained by the NUCC and its fields are items; the UB-04 is maintained by the NUBC and its fields are form locators. Candidates who memorize the form names but not the transaction names lose points in the billing sections. Chapter 1 §1.5, Chapter 26 §26.1.
The transaction pairs are worth memorizing as pairs, because the numbering is not intuitive: 270/271 eligibility inquiry and response; 276/277 claim status inquiry and response; 278 services review, which is prior authorization; 837 the claim; 835 the remittance; 834 enrollment; 820 premium payment. The trap is 278 versus 276/277: 276/277 asks where a submitted claim is; 278 asks permission for a service that has not happened. Chapter 24 §24.3, Chapter 27 §27.2.
The 277CA is an acknowledgment, not a status response, and exams use the similarity deliberately. Chapter 27 §27.6.
A 277 tells you a claim's status; an 835 tells you what was paid. A claim can show "finalized/denied" on a 277 and produce an 835 showing zero payment — the same event described by two transactions, and only the 835 carries the reason codes. Chapter 27 §27.6, Chapter 28 §28.4.
Item 24 clusters into four questions. Four modifiers fit on a line in 24D (more requires 99 plus item 19). Twelve diagnoses fit in item 21 and four pointers per line in 24E — the asymmetry is the question. 24F holds the charge, not the allowed amount, and candidates fresh from the pricing chapter reach for the wrong one. Units go in 24G. And the form has six service lines, so a longer claim becomes multiple claims. Chapter 25 §25.4, §25.5.
Pointers point letters, and only at what supports that line. Pointing all four positions at everything is not thoroughness — it is a claim that the venipuncture was performed for the hypertension. Chapter 25 §25.5.
Rejection is not denial, and the distinction is money. See I.15. Chapter 27 §27.7.

Review. Chapter 24 §24.3, §24.5 · Chapter 25 §25.4, §25.5, §25.7, §25.8 · Chapter 26 §26.1, §26.2, §26.3, §26.4, §26.5, §26.7 · Chapter 27 §27.2, §27.6, §27.7 · Appendix D for the field reference.

Self-check.

Name the transaction for each: verify coverage before the visit; request prior authorization; find out what happened to a submitted claim; receive the payment detail.
How many diagnoses fit in item 21, and how many may point to a single service line?
A hospital bills its outpatient department's facility resources. Which form, which transaction, which committee maintains it, and what are its fields called?
A claim comes back from the clearinghouse before the payer ever saw it. What is that called, and what are its appeal rights?

Answers: 1 — 270/271; 278; 276/277; 835. 2 — twelve; four. 3 — UB-04 (CMS-1450), 837I, the NUBC, form locators. 4 — a rejection; none, because there is no adjudication to appeal.

I.11 Domain 9 — Remittance, denials, appeals, and accounts receivable

What is tested. Reading a remittance; classifying a denial and deciding what to do with it; the appeal ladder; and the arithmetic a business office runs on. On a billing credential this is the heart of the exam.

The distinctions candidates miss.

The 837 goes out; the 835 comes back. They are the two ends of one transaction, and a surprising number of people confuse them under time pressure. The 835 is the transaction, ERA is the document it carries, SPR is the paper version. Chapter 28 §28.1.
The remittance advice goes to the provider; the explanation of benefits goes to the patient, and they say different things on purpose. Chapter 28 §28.2.
The group code carries the consequence. CO is a contractual obligation the patient does not owe; PR is patient responsibility; OA and PI are neither. Posting a CO amount as patient responsibility creates a bill nobody owes. Chapter 28 §28.3.
Denial items are scenario items, and the question is usually who should fix it or what the next step is — and the plausible answer is usually "the coder." Read the scenario for where the failure originated, not for where the denial landed. Coverage that terminated before the visit is an eligibility failure preventable at check-in, not a coding denial. A duplicate denial with no duplicate in your own system means look outward. Chapter 29 §29.2, §29.4.
A question offering "appeal it" for a front-end rejection is offering an action that does not exist. Chapter 29 §29.1.
Redetermination is first and belongs to the Medicare Administrative Contractor; reconsideration is second and belongs to the Qualified Independent Contractor. Alphabetical order survives exam pressure: rede- before recon-, MAC before QIC. The redetermination is the same contractor with different staff; the QIC is a different organization, so a stem describing "review by an independent contractor" is level two. The filing windows are not symmetrical — 120 days, then 180, then 60, 60, 60 — and the amount in controversy requirement starts at level three, so a question offering a dollar threshold for redetermination is testing whether you know there is none. Chapter 30 §30.6, §30.7.
Days in accounts receivable is total AR divided by average daily charges, and the stem's period is the trap: annual charges must be divided by 365 before anything else happens. The distractors are what you get by skipping that step or dividing by 12. Chapter 31 §31.3.
The aging buckets are conventions — 0–30, 31–60, 61–90, 91–120, 121+ — and the exam treats them as facts. That the honest content of a bucket depends on dating and netting choices is a working-life lesson layered on top of the exam answer, not a replacement for it. Chapter 31 §31.2, §31.3.
"Collection rate" unqualified means net on an exam, and means whatever the report writer wanted in real life. Chapter 23 §23.10, Chapter 31 §31.3.

And the theme this domain carries, stated in kind and not in numbers. Chapter 29 §29.8 is about prevention and Chapter 30 §30.1 is about deciding whether to appeal at all, and the book's fourth theme — a clean claim is a fast claim; preventing a denial upstream beats chasing money downstream — is the reason those two sections exist in that order. The assembly of what that is worth on a specific file belongs to Chapter 40 and is not performed here or anywhere else.

Review. Chapter 28 §28.1, §28.2, §28.3, §28.4, §28.5, §28.6 · Chapter 29 §29.1, §29.2, §29.4, §29.5, §29.6 · Chapter 30 §30.2, §30.5, §30.6, §30.7 · Chapter 31 §31.2, §31.3, §31.4, §31.9 · Chapter 32 §32.3, §32.4 · Appendix E for the denial codes.

Self-check.

A remittance line shows a CO adjustment and a PR amount. Which one may be billed to the patient?
A claim denies for a service the patient's plan does not cover. A second denies because filing was late. A third denies as a duplicate. Who owns each, and which are preventable?
Order the first two levels of Medicare appeal, name the decider at each, and give the filing window for each.
Annual charges are \$3,650,000 and total AR is \$400,000. Compute days in AR.
Which of these has appeal rights: a denial, or a rejection?

Answers: 1 — the PR amount only. 2 — coverage/benefit, and it is not winnable by argument; timely filing, owned by the process and almost never winnable — but check first whether the claim was rejected earlier and resubmitted late, which makes it a preventable administrative write-off; duplicate, and if there is no duplicate in your system, look outward (Chapter 29 §29.2). 3 — redetermination, the MAC, 120 days from the initial determination; reconsideration, the QIC, 180 days from the redetermination. 4 — 3,650,000 ÷ 365 = \$10,000 average daily charges; 400,000 ÷ 10,000 = 40 days. 5 — the denial; a rejection was never adjudicated.

I.12 Domain 10 — Facility coding: inpatient and outpatient

What is tested. Two prospective payment systems and the coding decisions that drive each. This is the center of a facility credential and a light conceptual touch on a professional one — check your content outline before you decide how many hours it gets.

The distinctions candidates miss — inpatient.

The arithmetic questions are gifts: relative weight times base rate, read from a supplied table. The traps are conceptual. Chapter 33 §33.5.
Severity tiers do not stack. Adding a second complication or comorbidity to a record that already carries a major one changes nothing. Chapter 33 §33.3.
A condition that developed after admission cannot be the principal diagnosis, however severe it became, and the answer choices will invite you to sequence it there. The principal diagnosis is the condition established after study to be chiefly responsible for the admission. Chapter 33 §33.2, §33.7.
Length of stay does not change the payment within the ordinary range — that is the entire meaning of "prospective." Chapter 33 §33.8.
Present on admission is its own indicator set (Y, N, U, W, and the exempt value), and it is what separates a condition the patient arrived with from one the hospital caused. Chapter 33 §33.7.
On ICD-10-PCS, the tested distinctions are root-operation boundaries: Excision versus Resection (a portion versus all of a body part — and the body part definitions do the work), Extirpation versus Excision (taking out foreign or solid matter versus cutting out the body part), Insertion versus Replacement (does the device take the body part's place?). And the other reliable stem gives you a procedure name in the operative report's header with documentation that contradicts it: code the objective documented, not the title. Chapter 33 §33.9.
Weights, the complication and comorbidity lists, and the group definitions change annually, effective October 1. Chapter 33 §33.5.

The distinctions candidates miss — outpatient.

The facility resources are reported on the UB-04 / 837I — and the professional claim's place of service still has to say it happened in a hospital outpatient department. That is the detail candidates drop. Chapter 34 §34.8, Chapter 25 §25.8.
Status indicators are read, not recalled. The exam supplies the letters and tests whether you know what they do: some package into another payment, some are discounted when multiple procedures are reported, some do not pay under the outpatient system at all. Chapter 34 §34.4.
Packaging means a line can pay \$0.00 by design. A coder trained on the professional claim, where every line has an allowed amount, reads a packaged line as a lost line. Chapter 34 §34.5.
Evaluation and management leveling splits between the two claims: national medical decision making rules on the professional side, and the hospital's own consistently applied criteria on the facility side. Chapter 34 §34.8.
Condition code 44 has requirements and a deadline, and §34.10 gives them as a checklist.
The update reflexes: the outpatient payment system annually; the outpatient code editor and the NCCI edits quarterly. Chapter 34 §34.7.

Review. Chapter 33 §33.2, §33.3, §33.4, §33.5, §33.7, §33.8, §33.9 · Chapter 34 §34.2, §34.3, §34.4, §34.5, §34.7, §34.8, §34.10 · Chapter 26 §26.3, §26.4, §26.7 for the form that carries all of it · Chapter 16 §16.9 for the one-encounter-two-claims structure.

Self-check.

A record already carries a major complication. A second, lesser comorbidity is added. What changes?
A patient is admitted with one condition and develops a more severe one on day two. Can the second be the principal diagnosis?
A facility outpatient line pays \$0.00. Name two entirely different reasons that could be correct.
Excision or Resection: the surgeon removes an entire lobe of the lung. Which root operation, and what makes the answer turn on something other than the word "entire"?

Answers: 1 — nothing; the tiers do not stack. 2 — no. 3 — it was packaged into another service's payment; or its status indicator is one that does not pay under the outpatient system. 4 — Resection, because the lung lobe is itself a body part in the classification — the body-part definitions decide, not the surgeon's phrasing.

I.13 Domain 11 — Specialty and risk-adjusted coding

What is tested. Not specialty expertise — the conventions that make a specialty different. And, separately, risk adjustment, which has its own credential and behaves unlike everything else in the book because the payment consequence is a population's score next year rather than this claim's allowed amount.

The distinctions candidates miss — specialty.

The trap in a specialty stem is almost always an included component listed as though it were a separate service. When a specialty scenario reads like a list, ask what the primary code already contains before you report anything from the list. Chapter 35 §35.1.
The recurring specialty stems are conventions stems: the catheterization with its components listed separately; the fracture scenario that turns on who is providing definitive care; the obstetric patient who transferred care partway through; the vaccine scenario missing its administration line; the anesthesia scenario that supplies room time and anesthesia time and hopes you use the wrong one. Chapter 35 §35.1.
The component split decides more specialty claims than any clinical fact. An electrocardiogram performed and interpreted by one entity is one code; an interpretation and report alone is a different one. Chapter 35 §35.2, Chapter 19 §19.1.

The distinctions candidates miss — risk adjustment.

The reset. A condition documented in detail two years ago and not since does not count for this year. Candidates who answer from clinical logic — "the condition still exists" — get it wrong; the question is about the collection period. Chapter 36 §36.6.
The hierarchy. Two severities of the same disease are reported and "both" is among the choices. One counts: the most severe. Chapter 36 §36.5.
The acceptable encounter. A condition supplied by a laboratory report, a problem list, or a nurse's telephone note cannot be reported. The discriminating fact is always the same: was there a face-to-face encounter with a diagnosing provider who assessed it? Chapter 36 §36.6, §36.7.
The linkage. The "with" convention, now with money on it — the presumption applies where the classification creates it and nowhere else, and the exam tests the boundary at least as often as the rule. Chapter 36 §36.7, Chapter 9 §9.7.
MEAT is the documentation standard: monitored, evaluated, assessed or addressed, treated. It is the book's first theme with an acronym attached. Chapter 36 §36.7.

Review. Chapter 35 §35.1, §35.2, §35.3, §35.4, §35.5, §35.7 · Chapter 36 §36.3, §36.4, §36.5, §36.6, §36.7, §36.9.

Self-check.

A stem lists catheter placement, angiography, and an intervention in one vessel. How many codes?
A patient's chronic condition was documented thoroughly two years ago and has not been addressed since. Does it count this year?
A problem list carries a condition that no note this year assesses. May it be reported for risk adjustment?
Expand MEAT and say which theme of this book it restates.

Answers: 1 — the intervention code includes the rest. 2 — no. 3 — no; a problem list is not an assessment at a face-to-face encounter. 4 — monitor, evaluate, assess/address, treat; the first theme, if it isn't documented, it didn't happen.

I.14 Domain 12 — Auditing and clinical documentation integrity

What is tested. On coding exams, lightly and in three predictable shapes. On an audit credential, as the whole subject.

The distinctions candidates miss.

Three reliable shapes on a coding exam. (1) A stem gives documentation and a billed code and asks what the auditor should do — and the trap answers are "query the provider" when the question is about scoring, or "downcode" when the service is not separately supported at any level. (2) Which authority governs a given finding: coverage to the national or local determination; bundling to the NCCI and its Policy Manual; documentation content to the record and the CPT guidelines. (3) The date rule — a claim is scored against the rules in force on the date of service. Chapter 37 §37.3.
Prepayment and postpayment review are not gentle and harsh versions of the same thing. Postpayment review takes money you already have. Prepayment review takes your calendar, your staff, and your cash cycle at once, and it does not end on a date you control. A claim under prepayment review does not deny; it suspends, pending an additional documentation request with a deadline printed on the letter — and a non-response is not neutral, it is a denial that counts against you as an error. Chapter 37 §37.4.
Precision and recall are engine-evaluation vocabulary, not code-set vocabulary. Learn them because you will be in the room when a tool is evaluated. Chapter 38 §38.9.
The query rules are the heavily tested half of Chapter 38 — see I.5, where they sit with the rest of the documentation material.

Review. Chapter 37 §37.2, §37.3, §37.4, §37.5, §37.6, §37.9 · Chapter 38 §38.1, §38.2, §38.3, §38.4.

Self-check.

An auditor finds a note that supports no level of the service billed. What is the finding, and why is "downcode it" wrong?
A claim from two years ago is being reviewed. Which year's rules apply?
Which authority settles a bundling finding, and which settles a coverage finding?

Answers: 1 — the service is not supported; downcoding presumes some level was earned and the record does not support that either (Chapter 37 §37.3). 2 — the rules in force on the date of service. 3 — the NCCI and its Policy Manual; the national or local coverage determination and its billing and coding article.

I.15 The high-yield distinctions

These are the pairs the whole book has worked to separate. If you are down to your last hour, this is the section. Each one has cost real practices real money, which is why each one is on an exam.

Denial versus rejection. A denial means the payer adjudicated the claim and declined to pay. A rejection means the claim never got adjudicated — it failed a front-end edit at the clearinghouse or the payer's gateway. A rejection has no appeal rights, because there is no decision to appeal; you correct it and resubmit, and the timely filing clock has been running the whole time. Chapter 29 §29.1 owns it; Chapter 27 §27.7 owns the pipeline half.

Charge versus allowed versus payment versus patient responsibility. Four numbers, never synonyms. Charge is what the provider bills; allowed is what the contract says the service is worth; payment is what the payer sends; patient responsibility is the part of the allowed amount the patient owes. The gap between charge and allowed is the contractual adjustment and nobody owes it. Chapter 1 §1.2.

Upcoding versus downcoding — and both are errors. Upcoding reports a service the documentation does not support. Downcoding reports less than the documentation supports. Downcoding is not the safe choice: it is inaccurate, it underpays the practice, it distorts quality and risk data, and it is not a defense. Chapter 5 §5.8.

Excludes1 versus Excludes2. Excludes1 means not coded here — the two conditions cannot be reported together. Excludes2 means not included here — the condition is not part of this code, so if the patient has both, report both. Excludes2 is permission, not prohibition. Chapter 8 §8.5.

First-listed versus principal diagnosis. First-listed is the outpatient concept: the condition chiefly responsible for the encounter, governed by Section IV. Principal is the inpatient concept: the condition established after study to have occasioned the admission, governed by Section II. Same idea, different settings, different rules — and an item that says "seen in the physician's office" and one that says "admitted" are asking different questions with the same words. Chapter 9 §9.3 and §9.4.

Modifier 25 versus modifier 57. 25 attaches a significant, separately identifiable evaluation and management service to a minor procedure — a 000 or 010-day global period. 57 identifies the evaluation and management service at which the decision for major surgery was made — a 090-day global. The global period decides, and candidates swap them. Chapter 14 §14.4 and §14.9.

Modifier 51 versus modifier 59. 51 says multiple procedures were performed at the same session and the payer should apply its multiple-procedure logic. 59 says this service was distinct — different session, site, lesion, incision, or injury — and is what overrides a bundling edit. They answer different questions and are never alternatives, and 59 yields to any more descriptive modifier. Chapter 14 §14.5 and §14.6.

Professional versus technical component. The professional component (modifier 26) is the physician's work: the interpretation and the report. The technical component (TC) is the equipment, the supplies, and the staff. One entity furnishing both reports the global service; two entities report their own halves, and nobody reports the global code. Chapter 14 §14.7, Chapter 19 §19.1.

CC versus MCC. A complication or comorbidity and a major complication or comorbidity are two tiers of severity that move a stay into a higher-paying group. They do not stack — a second condition at a lower tier adds nothing once the higher tier is present — and the lists are revised annually. Chapter 33 §33.3.

ICD-10-CM versus ICD-10-PCS. ICD-10-CM classifies diagnoses, in all settings, on every claim. ICD-10-PCS classifies inpatient hospital procedures only. A physician's procedure is never PCS, wherever it was performed. Chapter 33 §33.9 teaches the structure.

Medical necessity is a coverage concept, not a clinical one. A service can be entirely appropriate clinically and still not be medically necessary in the payer's sense, because "medically necessary" means meets this payer's published criteria for coverage. The physician's judgment and the coverage determination are two different documents answering two different questions, and conflating them produces both bad appeals and bad conversations with patients. Chapter 22 §22.1.

The facility claim versus the professional claim for one encounter. One visit to a hospital outpatient department produces two claims: the professional claim for the physician's work (CMS-1500 / 837P) and the facility claim for the room, the staff, the equipment, and the supplies (UB-04 / 837I). Two adjudications, two remittances, two patient balances, and a patient who received one service and two bills. Chapter 1 §1.5, Chapter 16 §16.9, Chapter 34 §34.8.

And the one that is a corollary rather than a pair: a rejection has no appeal rights. It is listed twice on purpose. A question offering "appeal it" for a front-end rejection is offering an action that does not exist, and that is the whole point of the item. Chapter 29 §29.1.

This book's position, and Chapter 39 §39.6's, is that the exams test whether you can find the answer, not whether you remember it. Which means the highest-yield hour in your last three weeks is not reading — it is timed retrieval, in the books you will actually sit with.

Six drills follow. Run each one against your own current code books or the encoder you will have, and time yourself with a phone in another room and a clock you can see. The times below are training targets, not exam parameters — your real per-item budget comes from your own exam's published item count and time limit, run through the arithmetic Chapter 39 §39.9 demonstrates.

Before you start, three ground rules.

Score two things, not one: did you get there, and did you get there by a route you could repeat? A right answer found by luck is a wrong answer that has not happened yet.
Every miss goes in the error log, with the reason column.
Do not look at the answer before you write yours down. Recognition is not retrieval.

   DRILL 1 -- THE TWO-STEP, COLD                          target 60 sec each
   ------------------------------------------------------------------------
   Ten diagnostic statements you write yourself, on index cards, from
   any chapter's worked charts. For each: MAIN TERM -> subterm -> code
   candidate -> VERIFY IN THE TABULAR -> read every instructional note
   above it.
   SCORE: the verification step is not optional and not scored separately.
   An answer found in the Index and not verified is a miss, even if the
   code is right. Chapter 8 8.1 is the rule this drill enforces.

   DRILL 2 -- FIND THE GUIDELINE                          target 45 sec
   ------------------------------------------------------------------------
   Someone reads you a rule; you find where it is written. Work from
   the list of guideline questions in Chapter 9: sequencing, the
   uncertain-diagnosis rule, signs and symptoms, the "with" convention,
   acute and chronic. Say the section aloud, then open to it.
   SCORE: knowing that a rule exists is worth nothing under a clock if
   you cannot land on it. This is the drill that makes tabs pay.

   DRILL 3 -- SECTION GUIDELINES, EVERY TIME              target 30 sec
   ------------------------------------------------------------------------
   Pick any CPT code at random. Before reading its descriptor, find and
   read the GUIDELINES at the front of its section and subsection.
   SCORE: Chapter 13 13.4 calls these the part everyone skips, and they
   set the meaning of every code beneath them. Ten repetitions of this
   drill changes how you read the book permanently.

   DRILL 4 -- THE FOUR-ANSWER ADJUDICATION                target 90 sec
   ------------------------------------------------------------------------
   Take any worked example in Parts II or III. Cover the answer. Write
   FOUR candidate codes yourself -- the right one and three near misses
   built from Chapter 39 39.6's list: wrong laterality, unspecified
   where specified is documented, correct code missing a required
   modifier, right family at the wrong level of the semicolon
   convention, right code for the WRONG SETTING.
   Then choose between your own four and say WHY in one sentence.
   SCORE: this is the single best transfer drill in this appendix,
   because writing the distractors teaches you to see them.

   DRILL 5 -- SETTING AND CLAIM FIRST                     target 15 sec
   ------------------------------------------------------------------------
   Twenty scenario openings, no clinical detail read. For each, name in
   three seconds: INPATIENT or OUTPATIENT, and PROFESSIONAL or FACILITY.
   SCORE: Chapter 39 39.2 says candidates who do this in the first three
   seconds do not make an entire class of error at all. Twenty
   repetitions is roughly five minutes and it is the cheapest points in
   this guide.

   DRILL 6 -- THE MODIFIER TABLE, FROM A STEM             target 60 sec
   ------------------------------------------------------------------------
   Ten one-line circumstances -- "the surgeon operated and sent the
   patient home to another practice for follow-up"; "the study was
   performed at the hospital and read by an outside radiologist";
   "the same test was repeated later the same day to obtain subsequent
   results." Name the modifier and the reason, then verify in
   Appendix B and in your own book's modifier appendix.
   SCORE: reason first, modifier second. A modifier chosen without a
   stated reason is the exact habit that produces an audit finding.

How to build a week around these. Two drills a night, alternating, in the twenty minutes before you stop. They are deliberately short: the point is frequency, not duration. Chapter 39 §39.7 puts timed practice in weeks ten and eleven and it is right to — accuracy first, then speed, never the reverse — but Drills 3 and 5 cost so little that they belong in week two.

And one drill you should not run. Do not practice by reading answer keys. It feels efficient and it trains recognition, which is the skill the exam is specifically designed not to reward.

I.17 What this guide cannot do, and where to go next

It cannot tell you what your exam contains. It can tell you what this book teaches, where each idea lives, and which distinctions have historically cost people points. The authority on your exam's domains and their weights is the current exam content outline; the authority on every parameter of sitting it is the current candidate handbook; and both belong to the organization administering the exam, not to any book.

It cannot make you fast. Retrieval speed is built by repetitions and there is no way around the volume. What this appendix can do is make sure the repetitions are aimed at the right material.

And it is not the plan. For the plan — the twelve weeks, the hours, the error log and the column that changes behavior, the two full-length simulations, the annotation discipline, the accommodations process, the exam-day clock, the three passes, the apprentice designation, and what happens after you pass — go back to Chapter 39, which owns all of it: §39.7 for preparation, §39.8 for your code books, §39.9 for the morning itself, §39.10 and §39.11 for what comes after.

For practice with the reasoning shown, go to Appendix H, which works coding scenarios start to finish — documentation in, codes out, the path taken, and the plausible wrong answer named and rejected. That is the form the exam's items take, and it is the closest thing in this book to sitting one.

For the reference layer while you drill: Appendix A for code-set structure, Appendix B for modifiers, Appendix D for the claim-form fields, Appendix E for denial codes, Appendix F for the federal law, Appendix G for documentation and query templates, and Appendix J for what the credential is for once you have it.

One last thing, and it is the reason the whole book is organized the way it is. The exam will end. The code sets will change on October 1 and on January 1 and four times a year after that, and the habits you build in these weeks — main term, subterm, verify in the Tabular, read the conventions, read the guidelines, be able to explain the path two years later — are the only part of this that does not expire. Study for the exam. Build for the desk. They are, almost entirely, the same work.

Appendix I — The Certification Exam Study Guide

I.1 How to use this guide — and the one rule that governs every page of it

The rule, stated first because it outranks everything else here

How the two documents fit together

The revision loop that actually works

What this guide is not

I.2 The domain map — this book, rearranged

I.3 Domain 1 — The revenue cycle and reimbursement

I.4 Domain 2 — Compliance and the regulatory frame

I.5 Domain 3 — The medical record and documentation

I.6 Domain 4 — ICD-10-CM

The structure of a code

The lookup discipline

The Official Guidelines

Body-system applications

The special chapters

I.7 Domain 5 — CPT

The structure of the book

Modifiers

Evaluation and management

Surgery

Radiology, pathology and laboratory, and medicine

I.8 Domain 6 — HCPCS Level II

I.9 Domain 7 — Edits, medical necessity, and pricing

I.10 Domain 8 — The claim forms and electronic submission

I.11 Domain 9 — Remittance, denials, appeals, and accounts receivable

I.12 Domain 10 — Facility coding: inpatient and outpatient

I.13 Domain 11 — Specialty and risk-adjusted coding

I.14 Domain 12 — Auditing and clinical documentation integrity

I.15 The high-yield distinctions

I.16 The navigation drills

I.17 What this guide cannot do, and where to go next