Chapter 23: Pricing the Service: RVUs, the Medicare Physician Fee Schedule, and the Chargemaster

"Every number on that remittance came from somewhere. Three of them came from an equation, one came from a contract, and one came from a decision nobody at the practice remembers making." — constructed

Overview

Chapter 2 showed you the money moving. Charge, allowed, contractual adjustment, patient responsibility, plan payment — the four-number structure that every claim in this book has used.

It did not tell you where any of the numbers came from.

This chapter does. And it closes the last of the book's open questions.

Since Chapter 2 you have been carrying Q6 — why did the practice charge \$185.00? The answer is here, and it is not the answer most people expect.

Three numbers, three sources, and they are genuinely different in kind:

   $96.52  ...  Medicare's allowed amount for 99214
                ► an EQUATION. Public inputs, reproducible,
                  and you can compute it yourself in a minute.

   $128.40 ...  Northfield Mutual's allowed amount
                ► a CONTRACT. Negotiated, private, and knowable
                  only by reading it.

   $185.00 ...  the practice's charge
                ► a DECISION. Made once, adjusted since, and
                  — this is the honest part — not derivable
                  from anything.

In this chapter, you will learn to:

• Compute a Medicare allowed amount from published inputs
• Say what each RVU component measures
• Apply geographic adjustment, and know what it does not adjust
• Explain the conversion factor and the annual fight
• Apply the site-of-service differential
• Build a fee schedule
• State the one rule that actually governs charge setting
• Describe a chargemaster and how it decays
• Compute gross and net collection rates — and say what each hides

23.1 Where \$128.40 comes from

Start at the end and work backward, because the destination explains the arithmetic.

Account 10-4471's line 1 allowed \$128.40. Where did it come from?

Not from an equation. \$128.40 is a contracted rate — a number in Northgate Family Medicine's agreement with Northfield Mutual, negotiated at some point by people who are not in this book.

So why does this chapter spend most of its length on Medicare's arithmetic?

Because the contract is written against it. Commercial contracts are overwhelmingly expressed as a percentage of the Medicare Physician Fee Schedule — or as a schedule built from one — and:

• You cannot evaluate a contract you cannot price. "112% of Medicare" is a number only if you can compute Medicare.
• You cannot detect an underpayment without knowing what was owed. Chapter 14's Case Study 2 and Chapter 18's Case Study 2 both turned on this.
• And the fee schedule is the closest thing to a public price list American health care has. It is free, it is complete, and it is the reference every other number in the system is described against.

So: \$128.40 is a contract. \$96.52 is the equation underneath it. The rest of this chapter is that equation, and then what the practice did with it.

23.2 The relative value unit and its three components

A relative value unit (RVU) expresses the resources required to provide a service relative to other services. It is a ratio, not a dollar amount.

Every code in the physician fee schedule carries three RVUs, and they measure three different things.

Three things worth knowing about the components.

The work RVU is the one physicians talk about, because it is the one that measures them. It is also frequently used inside organizations as a productivity measure — physician compensation is frequently expressed in dollars per work RVU — which means the work RVU has a life entirely outside claim payment.

The practice expense RVU comes in two flavors, and this is the whole of §23.5: a non-facility value, used when the practice supplied the overhead, and a facility value, used when a facility did.

And the malpractice RVU is small and it is real. It is usually the smallest of the three by a wide margin, and it varies more by specialty and geography than the others.

Where the numbers come from

RVUs are not measured. They are determined, through a process worth understanding because it explains a great deal about why the values look the way they do.

   A new or revised code is created (CPT Editorial Panel)
        ▼
   A specialty society SURVEYS its members about the work
        ▼
   A multispecialty committee — the RUC — reviews the survey
   and RECOMMENDS relative values to CMS
        ▼
   CMS accepts, modifies, or rejects the recommendation
        ▼
   The value appears in the proposed rule, then the final rule
        ▼
   PERIODIC REVIEW: codes are revisited, particularly those
   identified as potentially misvalued

Four things follow that a coder will otherwise find inexplicable.

The values are comparative judgments, not measurements. A work RVU expresses how a code's work compares to other codes' work, as assessed through survey and committee. There is no instrument that measures physician effort, and expecting the numbers to behave like measurements will mislead you.

Specialty societies participate structurally. The survey step means the societies representing the physicians who perform a service are the ones who characterize its work. This is transparent, it is intentional, and it is also exactly why the process is contested.

CMS is not obligated to accept the recommendation and does not always. The recommendation is advisory.

And codes get revalued. "Potentially misvalued" is a formal category, and services identified as such get revisited — which is how a code's RVUs can go down, and which is why §23.4's budget neutrality has something to redistribute.

🔢 Code It

Look up one code's RVUs.

The Medicare Physician Fee Schedule relative value file — the same free file that carried the six indicator columns in Chapters 17 through 19 — also carries the RVUs.

For 99214, the values this book uses (constructed teaching figures — verify current values):

text Work RVU ..................... 1.92 Practice expense (non-fac) ... 0.89 Practice expense (facility) .. lower — see 23.5 Malpractice RVU .............. 0.13

The file has one row per code and a column for each. Nothing about reading it is difficult; the only barrier is that most people have never opened it.

And notice what the three numbers say about a 99214: the physician's own work is roughly two thirds of it, overhead is most of the rest, and liability is a rounding error. That is a real statement about what an office visit costs to produce, and it is public.

23.3 Geographic practice cost indices

The same service costs different amounts to provide in different places, and the fee schedule adjusts for it.

A geographic practice cost index (GPCI) is a multiplier applied to each RVU component, reflecting relative costs in a specific payment locality.

There are three GPCIs — one per component — because the three components vary geographically in different ways:

• Work GPCI — reflects relative physician earnings across areas
• Practice expense GPCI — reflects rent, wages, and other overhead
• Malpractice GPCI — reflects liability premiums, which vary enormously by state

A GPCI of 1.000 means the locality is at the national average. Above 1.000 is more expensive; below is less.

Two things this book's frozen figures illustrate.

The values are close to 1.000 and they matter anyway. This book uses work 1.000, practice expense 1.008, malpractice 1.005 — a locality essentially at the national average. The adjustment moves the final number by pennies here. In a high-cost locality it moves it by dollars, on every claim, forever.

And GPCIs adjust cost, not value. A geographic adjustment does not say the service is worth more in one place. It says it costs more to provide, which is a different claim and a defensible one.

23.4 The conversion factor and the annual fight over it

RVUs are ratios. Something has to turn them into money.

The conversion factor is a single national dollar amount that converts adjusted RVUs into a payment. One number, for the entire physician fee schedule.

This is an unusual design and its consequences are large.

One number moves every payment. A change to the conversion factor changes the allowed amount for every code in the fee schedule, simultaneously, in the same proportion. There is no code-level negotiation; there is one lever.

Which is why it is fought over annually, publicly and loudly, by every physician organization in the country. The conversion factor is where the entire physician payment system's funding level is set, and it is set once a year.

Budget neutrality

And here is the mechanism that makes the fight unavoidable.

Budget neutrality requires that changes to RVUs — revaluing codes, adding codes, adjusting practice expense methodology — not increase total spending beyond a statutory threshold.

Which means the RVU pool is essentially fixed. If a set of codes is revalued upward, something else must come down, and the usual mechanism is a downward adjustment to the conversion factor.

Three consequences a coder will observe without necessarily understanding:

Revaluing one specialty's codes reduces everyone else's payment. This is why fee schedule changes generate the politics they do — it is structurally zero-sum.

A code whose RVUs went up may still pay less than it did last year, if the conversion factor came down more.

And "Medicare increased payment for X" and "Medicare cut payment overall" are frequently both true in the same year, describing the same rule.

The annual cycle, and why you should read the proposed rule

The fee schedule is set through rulemaking, on a predictable calendar.

   SUMMER ....... the PROPOSED RULE is published
                  ► RVU changes, conversion factor, policy changes
                  ► a COMMENT PERIOD opens

   AUTUMN ....... comments close; CMS reviews

   ~NOVEMBER .... the FINAL RULE is published
                  ► with responses to comments

   JANUARY 1 .... it takes effect

Two reasons this matters operationally.

You get months of warning and most organizations use none of it. A practice that reads the proposed rule in the summer knows by autumn what its payment will look like in January. A practice that does not, finds out from a remittance in February.

And the final rule's responses to comments explain the changes. Chapter 22 §22.4 said the same about LCDs and it is true here: the document that explains why is published alongside the document that does.

This is now the third public comment process this book has described — LCD development, NCD reconsideration, and the physician fee schedule rule. All three are open, all three are on a calendar, and almost nobody in the field participates in any of them.

🧮 Run the Numbers

The whole formula, once, with this book's frozen inputs.

(Constructed teaching figures — verify current values in the Medicare Physician Fee Schedule.)

99214, non-facility. Work RVU 1.92 · PE RVU 0.89 · MP RVU 0.13. GPCIs: work 1.000 · PE 1.008 · MP 1.005. Conversion factor \$32.7442.

Step 1 — adjust each component by its GPCI:

text work 1.92 × 1.000 = 1.92000 PE 0.89 × 1.008 = 0.89712 MP 0.13 × 1.005 = 0.13065 ───────── adjusted total RVUs = 2.94777

Step 2 — multiply by the conversion factor:

text 2.94777 × $32.7442 = 96.5224... → $96.52

\$96.52. That is the Medicare allowed amount for a 99214 in this locality.

You just priced a service from public inputs. Every number in that calculation is downloadable for free, and the arithmetic is two lines.

And now the comparison the rest of this chapter needs:

text Medicare allowed ........... $ 96.52 (an equation) Northfield Mutual allowed .. $128.40 (a contract) Northgate's charge ......... $185.00 (a decision)

23.5 The site-of-service differential

The same service pays a different amount depending on where it was performed, and the reason is entirely in the practice expense RVU.

When a service is performed in a facility — a hospital, an ambulatory surgical center — the facility supplied the overhead, and the physician's practice expense RVU is correspondingly lower.

When it is performed in a non-facility setting — a physician office — the practice supplied the overhead, and the practice expense RVU is higher.

The fee schedule file carries both values, side by side, for every code.

Three things follow.

The difference does not disappear — it moves. In the facility case, the facility bills separately for its own costs. Chapter 16 §16.1's two-claim structure is the same fact viewed from the claim side, and this is the payment mechanism underneath it.

The place of service code on the claim selects which value applies. Chapter 25 §25.8 covers the field. A wrong place of service code produces a wrong payment, in either direction, on a claim that is otherwise perfect — and it does not deny.

The codes themselves

A two-digit field, and the ones you will meet constantly:

Three notes.

11 versus 22 is the distinction that costs organizations the most. A practice acquired by a hospital and converted to a provider-based department — Chapter 16 §16.9 — changes place of service from 11 to 22, and the physician's payment drops because the facility now bills separately. An acquired practice still submitting 11 is reporting a setting that no longer exists.

19 versus 22 exists because off-campus and on-campus hospital outpatient departments are paid differently, which is a policy distinction rather than a clinical one.

And the telehealth codes are the least stable field in this table. Which place of service applies, and how it pays, has changed repeatedly. Verify current policy rather than carrying a habit.

And the total cost to the system is generally higher in the facility setting, because the facility's claim more than covers the physician's reduction. Chapter 16 §16.10's ED counterfactual measured this at 4.7×, and this section is why.

⚠️ Where Claims Die

A wrong place of service code is one of the purest silent errors in this book.

The claim is otherwise correct. Every code is right, every modifier is right, the diagnosis supports the service, and the edits pass. One two-digit field is wrong, and the payment is wrong.

It does not deny. The claim pays — at the facility rate when it should have paid the non-facility rate, or the reverse.

The office-to-facility direction is an underpayment, which accrues silently and expires under timely filing, exactly as in Chapter 14's Case Study 2.

The facility-to-office direction is an overpayment, which attaches Chapter 5 §5.1's sixty-day rule once identified.

And the detection is one query: compare the place of service on your claims against where the service was actually rendered, for one month. Practices with more than one site get this wrong routinely, and a practice that has recently added, moved, or acquired a location should check it deliberately.

23.6 Building a fee schedule from RVUs

The formula scales. Do it once per code and you have a fee schedule.

   FOR EACH CODE:

   [(work RVU × work GPCI)
      + (PE RVU × PE GPCI)
      + (MP RVU × MP GPCI)]  ×  CONVERSION FACTOR  =  allowed amount

Three practical notes for anyone who actually does this.

Choose facility or non-facility per code and per setting. A practice with both office and hospital work needs both columns.

The result is Medicare's allowed amount, which is what the arithmetic produces — not what any commercial payer pays.

And then the useful step: express your contracts against it.

🧮 Run the Numbers

What Northfield Mutual actually pays, expressed against Medicare.

(Constructed; every figure is Account 10-4471's, and the arithmetic is exact.)

Code	Medicare	Northfield	Northfield as % of Medicare
99214	96.52	128.40	133.03%
20610	63.28	78.60	124.21%
J1030	5.44	6.28	115.44%
36415	3.00	3.00	100.00%
Total	168.24	216.28	128.55%

Read the last column, and then read the last row.

The blended rate is 128.55% and NOT ONE LINE is at 128.55%.

This is the most important thing in the section. A practice that negotiated "roughly 128% of Medicare" and got this schedule is approximately right in aggregate and wrong on every single line.

Three consequences.

Your realized percentage depends on your service mix. A practice doing more E/M than this one earns above 128.55%; a practice doing more laboratory earns below it. The same contract is a different contract for a different practice.

A contract "at 128% of Medicare" and this schedule are not the same thing, and a practice comparing offers on the headline number is comparing something that does not exist.

And this is how you actually evaluate a contract: price your own top twenty codes by volume, compute the percentage for each, and weight it by your volume. That is your real rate. It is an afternoon's work in a spreadsheet and almost nobody does it.

23.7 How a practice sets its charges, and the myth of the high charge

Now Q6.

The one rule that actually matters

Before the history, the rule — because it is the only thing about charge setting that is not a matter of judgment.

A charge below a payer's allowed amount CAPS the payment at the charge.

Payers pay the LESSER of the billed charge and the allowed amount. A practice that charges \$90 for a service allowed at \$96.52 is paid **\$90.00**, and the \$6.52 is gone.

That is why charges are set above every contracted rate, and it is the entire functional requirement. Everything else about charge setting is policy.

And the corollary that people find surprising: a low charge is not generous. It does not help the patient — the patient's responsibility is computed from the allowed amount, not the charge. It only reduces what the payer pays, which is a transfer from the practice to the insurer and benefits nobody the practice cares about.

The myth of the high charge

And now the opposite error, which is more common and more publicly discussed.

A very high charge does not produce a higher payment from any contracted payer. The allowed amount governs. Charging \$500 for a service allowed at \$96.52 produces \$96.52 and a **\$403.48 contractual adjustment** — a number that looks impressive on a report and means nothing.

What a very high charge does do:

• It inflates the gross collection rate's denominator — §23.10, where this becomes a real problem
• It sets the amount a self-pay or out-of-network patient is asked for, which is a real consequence for a real person
• And it becomes the number quoted in every conversation about health care prices, because it is the only number that is public

The charge is not a price in the ordinary sense. It is a ceiling, an accounting convention, and — for the uninsured — an actual bill. Those three functions pull in different directions, which is why charge setting is genuinely hard and why most practices have not thought about it.

What a stated method looks like

A practice that wants a defensible answer needs a written method. In outline, and this is unglamorous:

   1. Price your own MEDICARE fee schedule (§23.6). Free, one afternoon.

   2. Identify your HIGHEST contracted rate per code, across all payers.

   3. Set the charge as a stated MULTIPLE of Medicare —
      chosen so it clears every contracted rate with margin.

   4. Apply the SAME multiple to every code.
      ► not "raise everything 3%"
      ► which preserves every existing distortion

   5. REVIEW ANNUALLY, when the conversion factor changes.

   6. WRITE IT DOWN, with a date and an owner.

Step 4 is the one that fixes the actual problem. Across-the-board percentage increases are why Northgate's four lines sit at 192%, 237%, 331%, and 467% of Medicare — the distortions were baked in long ago and every increase since has multiplied them. A single consistent multiple re-derived from the fee schedule flattens them in one pass.

And step 6 is what Chapter 16's Case Study 2 was about. A method that exists in someone's head is a method that will be gone in three years.

⚖️ Compliance Check

Charge uniformity is a real requirement and it is easy to violate without meaning to.

The general principle: charges should be uniform. The same service is charged the same amount regardless of the patient's payer, and the allowed amount — not the charge — is where payer differences live.

What that forbids is setting a different charge because of who is paying: a higher charge for a payer whose contract pays a percentage of billed charges, or a charge tuned per payer to maximize what each will allow.

What it does NOT forbid, and this trips people up: discounts to self-pay patients, financial assistance, and prompt-pay discounts, all of which are ordinary and appropriate when applied under a written, consistently applied policy. The charge stays uniform; what is collected can vary under a stated policy.

Two specific exposures:

Routinely waiving patient cost-sharing. Chapter 26 covers this at length; the short version is that a routine waiver of copayments or coinsurance is not generosity, it has been treated as an inducement, and it also means the amount actually charged is not the amount represented on the claim.

And charging an uninsured patient the full chargemaster amount while every payer pays a fraction of it. Legal in most circumstances and increasingly examined — and the mechanisms in §23.8 exist precisely because the outcome is hard to defend.

🗂️ The Encounter — Q6, ANSWERED

Why did Northgate charge \$185.00 for a 99214?

The honest answer, and it is the one worth having:

Nobody at the practice can derive it.

(Constructed history, and it is the ordinary one.) The charge originated years ago, from a fee-schedule reference product the practice subscribed to at the time. It has been increased since by across-the-board percentage adjustments — the kind that get applied to every line at once, in a meeting, in about ten minutes. Nobody has re-derived it against anything.

What it is not:

• Not a multiple of Medicare. \$185.00 ÷ \$96.52 = 191.67%. And the other lines are at 237.04%, 330.88%, and 466.67% — no consistent multiple exists.
• Not a computed figure of any kind. No one at Northgate holds a method that produces \$185.00.

What it does do — and this is the part that matters:

\$185.00 is above every contracted rate on this claim.** Northfield allows \$128.40; Medicare allows \$96.52. The charge does not cap anything.**

So the charge is performing its one essential function correctly, by accident, and that is genuinely the situation in most practices in the country.

Q6 is CLOSED, and the answer is that the number is a historical artifact that happens to work. A practice that wants a better answer needs a stated method — §23.9 — and the fact that it does not have one is a finding rather than a scandal.

23.8 The hospital chargemaster

A practice has a fee schedule. A hospital has a chargemaster, and it is a different kind of object.

The chargemaster, or charge description master (CDM), is the hospital's complete list of billable items and their charges — every procedure, supply, drug, room rate, and service the institution can bill for.

Three things that distinguish it from a physician fee schedule.

Scale. A chargemaster commonly contains tens of thousands of lines. A physician practice's fee schedule contains hundreds.

It includes items, not just services. Supplies, drugs, implants, and room-and-board rates all live here alongside procedure codes.

And it is a shared object with no single owner. Clinical departments request lines. Finance sets prices. Compliance worries about it. Information systems maintains the file. Chapter 16's Case Study 2 was a document nobody owned, and a chargemaster is the same problem at ten thousand times the scale.

How it decays

Four mechanisms, all ordinary:

Codes terminate and lines survive. Chapter 20 §20.1's quarterly HCPCS updates and CPT's annual ones retire codes; a chargemaster line pointing at a retired code denies on every claim until somebody connects the denials.

Items are added and never removed. A supply used for two years and discontinued leaves a line behind.

Prices are adjusted across the board. An across-the-board increase applied to every line preserves every existing distortion and adds a compounding one.

And lines accumulate that were never payable. Chapter 20 §20.10's A4550 surgical tray is the standing example — a correct code for a real item, generating a denial on every claim, forever, because somebody once found a code that matched.

Price transparency, and the chargemaster in public

The chargemaster used to be an internal document. It is not anymore.

Hospitals are subject to price transparency requirements obliging them to publish, publicly and without barriers, a machine-readable file of standard charges — including gross charges, payer-specific negotiated rates, discounted cash prices, and de-identified minimum and maximum negotiated charges — and a consumer-friendly display of a set of shoppable services.

Three consequences worth carrying.

The negotiated rates are public. §23.6's exercise — pricing a contract against Medicare — is now possible for other organizations' contracts, which is an entirely new fact about this industry and which changes what a negotiation looks like.

The chargemaster is now a public document with an internal history. Every decayed line, every dead item, and every never-payable charge in §23.8's list is published. An object that accreted for twenty years without an owner is now read by journalists, researchers, competitors, and occasionally patients.

And the quality of the files varies enormously, which is itself informative: a machine-readable file that is difficult to use is a statement about how well the underlying object is maintained.

And the patient who is asked to pay it

For an uninsured or out-of-network patient, the chargemaster is not an accounting convention. It is the bill.

Three mechanisms respond to that, and a coder should know they exist:

Financial assistance policies. Nonprofit hospitals operate under obligations to maintain and publicize them, and to limit what they charge patients eligible for assistance.

Discounted cash prices, which the transparency requirements oblige hospitals to publish and which are frequently far below the gross charge.

And good faith estimates. Under the federal surprise billing framework, uninsured and self-pay patients are entitled to a good faith estimate of expected charges before scheduled care.

📞 On the Phone

"Why is this three hundred dollars? My friend's insurance paid ninety."

The caller has understood the system correctly and is asking the right question. Do not treat this as a complaint to be deflected.

What works: "You're right that those are different numbers, and I'll explain why. The three hundred is our standard charge — it's the same for everyone, and insurance companies pay a discounted rate we've agreed to with them. Because you don't have coverage, you're being asked for the full charge, and that isn't a good outcome for you. We have a discounted self-pay rate and a financial assistance application, and I'd like to go through both with you."

Three things that does. It confirms the caller's observation rather than denying it. It explains the mechanism without jargon. And it offers the two things that actually help, which exist and which most patients do not know about.

What does not work: "That's our standard charge." True, unhelpful, and it leaves a person paying several times what any insurer pays for the same service — which is the outcome the discounted cash price and the assistance policy exist to prevent.

And know your own organization's numbers before this call. The self-pay discount, the assistance thresholds, and the good faith estimate process. A billing office that cannot answer this call is the reason the question is asked angrily the second time.

23.9 Charge capture and where charges are lost

Everything above assumes the charge got onto a claim. A meaningful share do not.

Charge capture is the process by which a service performed becomes a charge on an account.

Five places charges are lost, and they are worth knowing as a list because each has a different fix:

Services performed and never entered. The most common and the hardest to detect — there is no record of the thing that did not happen.

Services entered in one system and never interfaced to another. Chapter 16 §16.2's office-visit-plus- admission problem is this, and so is any two-system workflow.

Charges entered after the billing cycle closed, which either post late or drop.

Charges entered against a closed or wrong account.

And charges that were correctly captured and then removed by an edit or a rule — which is the one that looks like a control and is sometimes a leak.

⚠️ Where Claims Die

Lost charges are the only failure in this book with NO evidence at all.

A denial leaves a remittance. An underpayment leaves a payment. A charge that was never entered leaves nothing — no claim, no denial, no work queue item, no line on any report. You cannot reconcile against a record that does not exist.

Which means detection has to come from OUTSIDE the billing system, by comparing it to something that does have a record:

• The schedule — patients seen versus accounts with charges
• The medical record — notes signed versus charges posted
• Clinical logs — operative logs, infusion logs, procedure logs versus billed procedures
• Supply and pharmacy usage — Chapter 20's Case Study 1 reconciliation, run in the other direction

Every one of those comparisons is available and most organizations run none of them. A signed note with no charge is the single highest-yield query in this chapter, and it is one join.

Running the reconciliation

Four comparisons, in ascending order of effort and descending order of how many organizations do them.

Schedule to charges. For a date, every patient with a kept appointment should have an account with charges. Exceptions are a short list and each one has a reason — a no-show recorded late, a visit converted to a nurse encounter, a patient seen and not billed deliberately. A list that is not short is a finding.

Signed notes to charges. A signed note asserts a service was performed. A signed note with no charge is either a lost charge or a service somebody chose not to bill, and both are worth knowing about. This is one join and it is the query to build first.

Clinical logs to charges. Operative logs, infusion logs, endoscopy logs, therapy schedules. These are kept for clinical and regulatory reasons and they are independent of billing, which is exactly what makes them useful. Chapter 19's Case Study 2 lived in one of these.

And supply or pharmacy usage to charges. Chapter 20's Case Study 1 ran this comparison to find an overbilling. Run in the other direction it finds implants, devices, and drugs that were used and never charged — which in a facility is the largest single category of lost charge revenue.

🎓 Exam Watch

The RVU formula is a guaranteed exam item and it is always the same shape: three RVUs, three GPCIs, a conversion factor, and an answer to the cent.

Do it in two steps, every time. Adjust each component by its own GPCI, sum, then multiply by the conversion factor. Candidates who sum the RVUs first and apply one GPCI get a wrong answer that looks plausible.

Three related items that travel with it:

"Facility or non-facility?" — the question is who supplied the overhead, and the answer selects the practice expense RVU.

"The charge is \$90 and the allowed amount is \$96.52. What is paid?" — \$90. Payers pay the lesser of billed and allowed, and this is the single most tested fact in the chapter.

And gross versus net collection rate. Gross uses charges as the denominator; net uses what you were entitled to collect. Exams ask which one is meaningful, and the answer is net.

23.10 Gross and net collection rates, and what each one hides

Two ratios, both universally reported, and they measure completely different things.

GROSS collection rate = payments ÷ CHARGES

NET collection rate = payments ÷ (charges − contractual adjustments), i.e. payments ÷ what you were entitled to collect

What the gross rate hides

Almost everything.

Its denominator is charges, and §23.7 established that charges are arbitrary. A practice that doubles its charges halves its gross collection rate without any change in performance whatsoever.

Which means the gross rate is not comparable between organizations, is not comparable across time if charges changed, and does not measure collection performance at all. It measures the relationship between two numbers, one of which is a historical artifact.

It is reported constantly anyway, because it is easy to compute.

What the net rate measures — and what it still hides

The net collection rate is the real one. Its denominator is what you were entitled to collect after contractual adjustments, so it measures how much of your legitimate revenue you actually got.

But it hides two things, and they are important.

It cannot see charges that were never entered. §23.9's lost charges are absent from both the numerator and the denominator. A practice that loses 5% of its charges to capture failures can report a 99% net collection rate, and both numbers are correct.

And it cannot see underpayments that were posted as contractual adjustments. This is the subtle one. If a payer underpays and your system writes the difference off as contractual, the underpayment leaves the denominator along with the legitimate adjustment. The net collection rate rises.

A silent underpayment IMPROVES your net collection rate.

Chapter 14's Case Study 2 — four years of bilateral underpayment — would have made that practice's net collection rate look better every year it continued.

Which is why Chapter 28 §28.8's underpayment comparison exists, and why it has been promised in five chapters: it is the only measurement that sees what these two ratios cannot.

And it is worth naming what those two ratios are actually for, because they are not useless.

The net collection rate is a good measure of one thing: how well you convert entitled revenue into received revenue. Denials worked, appeals won, timely filing met, patient balances collected, accounts not written off carelessly. That is a real and important thing to measure, and a practice whose net rate is falling has a real problem.

What it cannot do is tell you whether the entitlement was right. That question — did we bill everything, did we bill it correctly, were we paid what the contract required — is answered by charge capture reconciliation (§23.9), coding audit (Chapter 37), and underpayment comparison (Chapter 28 §28.8). Three different measurements, three different failures, and none of them is a collection rate.

The mistake is not reporting the net collection rate. It is reporting it as though it were a report card for the whole revenue cycle.

🔍 Check Your Understanding

A practice reports a net collection rate of 98.5% — excellent by any benchmark.

Name three serious problems that are entirely consistent with that number.

Answer:

Lost charges. Services performed and never entered are in neither the numerator nor the denominator. The practice could be losing 5% of its revenue at capture and still report 98.5%.

Underpayments posted as contractual adjustments. Every dollar underpaid and written off as contractual leaves the denominator, which raises the ratio. The worse this problem is, the better the number looks.

Undercoding. Chapter 15's Case Study 1 — a systematic downward shift of one level — produces claims that are correctly coded for what they say, adjudicated correctly, and paid in full. Collection performance is flawless. The practice is simply billing for less than it did.

What all three have in common: the net collection rate measures how well you collect what you billed. It says nothing about whether you billed everything, billed it correctly, or were paid what the contract required. Three different questions, and the ratio answers one of them.

23.11 The three numbers, together

The chapter's whole content, in one place.

Three things to carry.

Only one of them is derivable, and it is the one nobody looks up. The Medicare arithmetic is free, public, and two lines, and it is the reference every other number is described against.

The contract is knowable and unknown. Northgate has the contract. Nobody there has priced it against Medicare per code, which is why the 128.55% blended rate in §23.6 was a discovery rather than a fact on file.

And the charge is doing its job by accident. \$185.00 is above every contracted rate, so it caps nothing. That is the requirement, and it is satisfied for no reason.

The gap between "we have a contract" and "we know what our contract pays" is the widest gap in this chapter, and closing it is an afternoon.

And this is where Part IV ends

Part IV asked one question in four different ways: what decides whether a claim gets paid, and how much?

Chapter 20 said the code has to exist and be the right kind of thing, and that a code existing is not a code being payable — four reasons, three settled before the claim goes out.

Chapter 21 said the combination has to survive an edit file that is public, free, and quarterly, and that overriding an edit requires documentation already written and already signed.

Chapter 22 said the service has to be one the payer agreed to cover for this patient — a coverage question, not a clinical one — and that the first thing to determine about any necessity denial is whether the claim was coded wrong or never going to be covered.

And Chapter 23 says the amount comes from three different kinds of place: an equation you can compute, a contract you have to read, and a decision nobody remembers making.

Part V builds the claim. Everything in Part IV has been about what the claim has to survive. Chapter 24 puts it on a form.

Summary

\$128.40 is a CONTRACT. Commercial rates are overwhelmingly expressed against the Medicare Physician Fee Schedule, which is why the arithmetic matters even when the arithmetic is not what paid you.

An RVU is a ratio. Work — the physician's own time, skill, effort, judgment, and stress. Practice expense — overhead, with separate facility and non-facility values. Malpractice — liability, the smallest and the most geographically variable.

RVUs are determined, not measured — a specialty society survey, a multispecialty committee recommendation to CMS, and CMS's acceptance, modification, or rejection. They are comparative judgments, specialty societies participate structurally, and codes get revalued, which is what budget neutrality has to redistribute.

GPCIs adjust each component for locality. 1.000 is the national average. They adjust cost, not value.

The conversion factor is a single national dollar amount, which is why one number moves every payment and why it is fought over annually. Budget neutrality makes the RVU pool essentially fixed — revaluing one specialty's codes reduces everyone else's payment, and a code whose RVUs rose may still pay less.

THE FORMULA

[(work × work GPCI) + (PE × PE GPCI) + (MP × MP GPCI)] × CONVERSION FACTOR

99214: (1.92 × 1.000) + (0.89 × 1.008) + (0.13 × 1.005) = 2.94777 × \$32.7442** = **\$96.52

The fee schedule is set through annual rulemaking — a proposed rule in summer with a comment period, a final rule around November, effective January 1. Months of warning, and most organizations use none of it. (The third open public process in this book, after LCD development and NCD reconsideration.)

The site-of-service differential lives entirely in the practice expense RVU. Facility pays the physician less because the facility supplied the overhead and bills separately. The place of service code selects which value applies, and a wrong one produces a wrong payment that does not deny. 11 office · 19 off-campus outpatient hospital · 21 inpatient · 22 on-campus outpatient hospital · 23 ED · 24 ASC. 11 versus 22 is the one that costs acquired practices the most.

Expressing a contract against Medicare is the useful step, and Account 10-4471's schedule shows why: the blended rate is 128.55% and no single line is at 128.55%. Your realized percentage depends on your service mix.

THE ONE CHARGE-SETTING RULE

A charge below a payer's allowed amount CAPS the payment at the charge. Payers pay the lesser of billed and allowed. Everything else is policy.

A low charge is not generous — the patient's share comes from the allowed amount. And a high charge does not increase payment from any contracted payer; it inflates the gross collection rate, and it sets what an uninsured patient is asked for.

Q6 is CLOSED: nobody at Northgate can derive \$185.00. Not a multiple of Medicare (191.67%, against 237%, 331%, and 467% on the other lines). A historical artifact, adjusted across the board, that happens to sit above every contracted rate — which is the one thing it has to do.

A stated charge method: price your Medicare schedule → find your highest contracted rate per code → set a single multiple of Medicare that clears it → apply the same multiple to every code → review annually → write it down with a date and an owner. "Raise everything 3%" preserves every existing distortion, which is why Northgate's four lines sit at 192%, 237%, 331%, and 467%.

Charges should be UNIFORM — the same service, the same charge, regardless of payer. Self-pay discounts, financial assistance, and prompt-pay discounts are ordinary under a written, consistently applied policy. Routinely waiving patient cost-sharing is not.

A chargemaster is tens of thousands of lines, includes items as well as services, and has no single owner. It decays through terminated codes, dead items, across-the-board adjustments, and lines that were never payable — A4550 being the standing example.

And it is public now. Price transparency requirements oblige hospitals to publish a machine-readable file of standard charges — including payer-specific negotiated rates — plus a consumer display of shoppable services. Other organizations' contracts can now be priced against Medicare the way §23.6 priced Northgate's. For uninsured and out-of-network patients the chargemaster is the bill, which is what financial assistance policies, discounted cash prices, and good faith estimates exist to address.

Charge capture loses services never entered, services never interfaced, late charges, wrong-account charges, and charges removed by rules. Lost charges are the only failure in this book with no evidence at all — detection must come from outside the billing system, and a signed note with no charge is the highest-yield query in this chapter.

Gross collection rate = payments ÷ charges, and its denominator is arbitrary — doubling charges halves it. Net collection rate = payments ÷ what you were entitled to collect, and it is the real one.

But a SILENT UNDERPAYMENT IMPROVES your net collection rate,

because the shortfall leaves the denominator as a contractual adjustment. And lost charges and undercoding are invisible to both ratios.

Key Terms

Relative value unit (RVU) — a measure of the resources required to provide a service, relative to other services. A ratio, not a dollar amount. (Ch.23)

Work RVU — the component measuring the physician's time, technical skill, physical and mental effort, judgment, and stress. Frequently used as a productivity measure independent of payment. (Ch.23)

Practice expense RVU — the overhead component, published in separate facility and non-facility values. (Ch.23)

Malpractice RVU — the professional liability component; typically the smallest and the most geographically variable. (Ch.23)

Geographic practice cost index (GPCI) — a locality multiplier applied to each RVU component, adjusting for relative cost. 1.000 is the national average. (Ch.23)

Conversion factor — the single national dollar amount converting adjusted RVUs into a payment. (Ch.23)

Medicare Physician Fee Schedule — the schedule of allowed amounts produced by the RVU, GPCI, and conversion factor arithmetic. (Ch.23)

Budget neutrality — the requirement that RVU changes not increase total spending beyond a statutory threshold, making the pool essentially fixed. (Ch.23)

Site of service differential — the payment difference produced by the separate facility and non-facility practice expense RVUs. (Ch.23)

Chargemaster (CDM) — a hospital's complete list of billable items and their charges, commonly tens of thousands of lines. (Ch.23)

Charge capture — the process by which a service performed becomes a charge on an account. (Ch.23)

Gross collection rate — payments divided by charges. Its denominator is arbitrary, so it does not measure collection performance. (Ch.23)

Net collection rate — payments divided by what the organization was entitled to collect after contractual adjustments. (Ch.23)

Percentage of Medicare — the common expression of a commercial contract's rates against the Medicare Physician Fee Schedule. The realized percentage depends on service mix. (Ch.23)

Potentially misvalued code — a formal category of codes identified for revaluation, through which a code's RVUs may go down. (Ch.23)

Place of service code — the two-digit claim field identifying where a service was furnished, which selects the facility or non-facility practice expense value. (Ch.23)

Price transparency requirements — obligations on hospitals to publish a machine-readable file of standard charges, including payer-specific negotiated rates, and a consumer display of shoppable services. (Ch.23)

Good faith estimate — the estimate of expected charges to which uninsured and self-pay patients are entitled before scheduled care. (Ch.23)

Charge uniformity — the principle that the same service carries the same charge regardless of payer, with payer differences living in the allowed amount rather than the charge. (Ch.23)

Spaced Review

1. Name the three RVU components and say what each measures. Which one has two published values, and why?

2. Compute the Medicare allowed amount for a code with work 1.92, PE 0.89, MP 0.13, GPCIs 1.000 / 1.008 / 1.005, and a conversion factor of \$32.7442. Show both steps.

3. What does budget neutrality do to the RVU pool? Explain how a code's RVUs can rise while its payment falls.

4. Where does the site-of-service differential live, and who receives the difference? (Chapter 16) Which chapter measured the total-cost consequence, and at what multiple?

5. A wrong place of service code produces a wrong payment. Why is this among the purest silent errors in the book? Give the detection query.

6. Account 10-4471's blended commercial rate is 128.55% and no line is at 128.55%. State two consequences.

7. State the one rule that governs charge setting. Then explain why a low charge is not generous and a high charge does not increase payment.

8. (Q6) Why did Northgate charge \$185.00? What is the honest answer, and what is the one thing the charge is doing correctly?

9. Name four ways a chargemaster decays. (Chapter 20) Which one does A4550 illustrate?

10. A practice reports a 98.5% net collection rate. Name three serious problems consistent with that number, and say what the ratio actually measures.

11. Describe how an RVU is established. Why does the chapter say the values are "comparative judgments, not measurements," and what follows from that?

12. Name the four stages of the annual fee schedule rulemaking cycle and when each occurs. (Chapter 22) Name the other two open public processes this book has described.

13. A practice is acquired by a hospital and converted to a provider-based department. What happens to the place of service code, and what happens to the physician's payment?

14. State the six steps of a defensible charge-setting method. Which step fixes the distortion that across-the-board increases create, and why?

15. Charges should be uniform. Name three things that are nevertheless permitted, and one that is not.

16. What must a hospital publish under price transparency requirements? Name one thing that becomes possible as a result that was not possible before.