When do you split an echo into professional (26) and technical (TC) components?

When does an echo actually split into 26 and TC?

An echo splits when two different entities furnish the two halves of it. Modifier 26 covers the physician work: supervision, interpretation and the written report. Modifier TC covers everything that produced the images: machine, sonographer, supplies, room. When one entity does both, there is no modifier and the global code stands alone.

The operative question is not who signed the lease on the scanner. It is where the patient was lying when the images were recorded. Chapter 13 of the Medicare Claims Processing Manual is direct about the consequence: contractors may not pay the technical component of a diagnostic service furnished to a hospital patient, nor for skilled nursing facility inpatients during a Part A covered stay. A practice cannot bill globally for a hospital patient even if it wanted to, because half of that global code is not payable to it.

Why equipment ownership is the wrong test

Most guidance reduces this to equipment ownership, and that test breaks in three common cardiology situations. A group that owns its machine and converts to provider-based status must switch every study to modifier 26 while nothing about the equipment changes. A mobile echo vendor brings a machine into the office, so the practice owns nothing and may still bill. An independent diagnostic testing facility owns the equipment and bills under its own enrollment rules. Ownership is a fact about the balance sheet. The modifier is a fact about the encounter.

Modifiers 26 and TC are never both appended to the same code on the same line, and a purchased technical component goes on its own line of service.

Which echo codes can carry modifier 26 or TC at all?

Not all of them, and this is where a correct understanding of the rule still produces a denied claim. Every code in the physician fee schedule carries a PC/TC indicator deciding which modifiers are legal on it. Indicator 1 means both components and both modifiers. Indicator 6 accepts 26 but not TC. Indicator 2 marks professional-only codes that stand alone, 3 marks technical-only codes, and 4 marks global-only codes. On 2, 3 and 4 the modifiers are not merely unnecessary, they are invalid, and the line is rejected.

Cardiology already has a familiar example outside echo. CPT 93000 and CPT 93010 are both cited by Novitas as codes that may not take modifier 26: one is global-only, the other is already the professional component. Nobody appends 26 to a 93010. The same logic governs part of the echo family, and far fewer people know it.

How to check an echo code PC/TC indicator this quarter

The file carrying these indicators is republished quarterly, not annually, so a code checked in January may read differently for an October date of service. Four steps, about ten minutes a quarter:

1. Open the CMS physician fee schedule relative value file for the quarter that contains the date of service, not the current quarter.
2. Find each echo code on your schedule and read its PC/TC indicator column.
3. Read its multiple procedure indicator in the same row. Indicator 6 flags a diagnostic cardiovascular service subject to the technical component reduction covered further down this page.
4. Diff the result against last quarter and correct any code whose indicator moved before the next claim goes out.

This is ordinary coding maintenance done against the current file rather than against a printed cheat sheet, and it is the single cheapest control on this whole topic.

The transesophageal codes where modifier 26 is an automatic denial

Transesophageal echo is already split at the code level, so the modifier that is correct everywhere else is wrong here. The family separates the work into one code covering probe placement, image acquisition, interpretation and report together, one covering probe placement alone, and one covering image acquisition, interpretation and report alone. The congenital set mirrors that structure, and a further code covers TEE for monitoring.

A biller who learns the universal rule and applies it to TEE appends modifier 26 to a code that is already professional-only. In our audit that error appeared on 9% of transesophageal claims. It is not an appeal. It is a corrected claim.

Which place of service goes on the professional claim?

This is where most split echo claims quietly go wrong, because the intuitive answer is wrong. CMS Change Request 7631, effective April 1, 2013, says the place of service reported by the physician is the setting in which the beneficiary received the technical component, not the location the physician was sitting in. The example CMS gives is exactly the echo scenario: a beneficiary has imaging at an outpatient hospital, the physician reads from the office, and the physician claim carries place of service 22.

That answers the question every group asks about after-hours and remote reads. It does not matter whether the cardiologist read from home, a second office or another state. The patient was in the hospital outpatient department, so the claim says 22.

In our audit, 11% of split echo claims carried a place of service that did not match where the images were recorded. Most pay at the wrong rate rather than denying, which is why they survive for years. Catching them at intake is front-end verification work rather than a coding project.

What changes when the practice goes provider-based

A group acquired by a health system keeps the same office, machine, sonographer and cardiologist, and still has to change how it bills every echo. The office is now a hospital outpatient department: global stops, modifier 26 starts, place of service moves from 11 to 19 or 22, and the facility side goes out institutionally. Miss the switch and you bill global for months, then face recoupment. That transition sits in the cardiology revenue cycle rather than in any one modifier.

What does the split actually pay? A worked example

Take a four-cardiologist group collecting $90,000 a month and running 190 complete echos: 120 in its own office on its own equipment, 70 at the hospital outpatient department and read by the group. Assume a local allowed amount of $228 global, $170 technical and $58 professional.

The global study is allowed $228, the technical component $170 and the professional component $58, so the professional half is roughly a quarter of the whole.

Now run the two errors. Bill all 190 globally and the 70 hospital studies go out against a technical component the hospital already billed, so they deny as duplicates. Correctly billed they are worth $58 each, $4,060 a month and $48,720 a year sitting in denial until somebody reworks them. Across the practices we review, 19% of denied claims were never reworked or appealed, turning $9,257 of that year into a write-off.

The second error is expensive precisely because it never denies. Append modifier 26 to office studies performed on your own equipment and you forfeit the $170 technical component on each. At 8% of 120 office studies a month, that is $1,632 a month and $19,584 a year. The claims pay, the remittance looks normal, nothing flags it. Finding it means comparing modifier against place of service on paid claims, the same blind spot that leaves underpayments against contracted rates on 7.8% of paid claims where the average underpaid claim was short by $38.

How do Doppler add-ons and stress echo change the answer?

The complete transthoracic study with spectral and color flow Doppler already contains the Doppler work, so reporting the Doppler add-on codes alongside it is unbundling rather than a component decision. The add-ons belong with the complete study reported without Doppler, and the add-on modifier has to match the base code. A base carrying 26 with an unmodified add-on is an inconsistent claim and is read as one.

Stress echo has its own trap. One stress echo code includes supervision, tracings and the report of the stress test; the other does not. Adding the separate stress test code alongside the version that already includes it is a duplicate, and in our audit it appeared on 6% of stress echo claims. Where the stress test is genuinely furnished separately, the answer is the component codes, not a modifier, because that family already splits itself the way TEE does.

What does MPPR take off the technical component?

For a busy echo lab this is the largest dollar consequence of the component decision. Codes carrying multiple procedure indicator 6 in the fee schedule are subject to it. The technical component with the highest non-facility total relative value on that date is paid in full, and each subsequent technical component is paid at 75 percent, a 25 percent reduction. The professional component is not reduced, and the reduction does not apply where the claim carries modifier 26 only.

On one patient, one date of service and two technical components, a $170 study and a $140 study pay $170 and $105 rather than $170 and $140, so the day total moves from $310 to $275.

A practice billing only professional components never sees this reduction. A lab reading remittances against undiscounted rates logs every second technical component as an underpayment and opens appeals that cannot win.

What goes wrong most often on split echo claims?

Ranked by how often we see them, the errors are: a hospital-performed study billed globally at 16% of hospital-performed echo studies; a place of service that does not match where the images were recorded at 11% of split echo claims; modifier 26 on a transesophageal code that cannot accept it at 9% of transesophageal claims; a Doppler add-on whose modifier does not match its base code at 7% of add-on lines; and the stress test code duplicated onto the stress echo code that already contains it at 6% of stress echo claims.

None of these is the biggest denial category in cardiology, and that is the point. They hide underneath the categories that are.

Across the Luxen claim audit, coding and modifier errors caused 21% of denials, prior authorization was missing on 13% of advanced cardiac imaging claims, device interrogation and remote monitoring claims were denied for frequency at 9%, underpayments against contracted rates appeared on 7.8% of paid claims, and professional and technical component errors appeared on 7% of in-office diagnostic claims. Component errors are the smallest bar on that chart and the one nobody owns, which is why cardiology practices carried a median 47 days in AR while these lines aged. Working them back is denial and AR recovery work, and appeals we file were overturned 68% of the time once somebody actually filed them.

Should echo component billing be handled in house or outsourced?

In house works when one named person owns the quarterly indicator check and reads place of service against modifier on paid claims. That is rarer than it sounds. Fully loaded in-house billing cost 7.9% of collections for practices under $2M in our billing reviews, against an outsourced range of 3% to 6% of collections.

Outsourcing is one option rather than the answer. It earns its keep when a group converts to provider-based status, adds a hospital reading contract, or loses the biller who understood the echo schedule, and it is the wrong purchase when the process already works. Ask any partner whether they check the fee schedule quarterly and whether they will show modifier-against-place-of-service reporting on paid claims, not just denials, which is the ground covered on how to evaluate a medical billing company. Worth knowing first: 44% of practice managers could not name the fee basis in their current billing contract.

Cardiology billing covers the rest of the code set, and full-service medical billing works inside the system you already use. Across our client practices, median days in AR dropped from 54 to 33 within 120 days. To have somebody read your last quarter of echo lines against place of service, a free billing review is where that starts.

Global, modifier 26 or modifier TC: which one does this echo get?

The three outcomes are decided by two facts: where the images were recorded, and whether the code has two components at all. Everything else follows.

One line worth keeping: global and modifier 26 fail in opposite directions, and only one of the two failures ever shows up on a denial report.

Cardiology

The whole question lives here, and the hard part is routing rather than coding. A group with an office lab and a hospital reading contract runs two billing paths through one worklist, and the sorting has to happen at scheduling rather than at charge entry. Tagging each study with its performing location at booking stops most of this at the source. Add the authorization layer on advanced imaging, missing on 13% of advanced cardiac imaging claims in our audit, and the front end does more work on an echo than the billing office ever will. Cardiology billing covers the rest of the schedule.

Anesthesia

Intraoperative TEE is where two specialties bill around one probe. The anesthesiologist who places the probe and the cardiologist who interprets are reporting different codes, not two modifiers on one code, and a monitoring TEE is a separate code from a diagnostic one. The common failure is an anesthesia group reporting the combined diagnostic code when it furnished only part of the service, or appending modifier 26 to a code that is already interpretation-only. Anesthesia time is miscalculated on 8% of cases in our audit, so the TEE line is rarely the only thing to check. See anesthesia billing.

Internal medicine and primary care

An internal medicine group with an in-office echo machine bills globally and usually correctly, right up until it refers a patient out and the read comes back to its own cardiologist. That single study is a modifier 26 claim with the facility place of service, and it will be the only one in the batch. The other pattern is assuming that ordering a study confers billing rights: it does not, and the ordering work is absorbed into the office visit. Internal medicine billing carries the wider visit side.

Radiology and imaging

Radiology groups and independent diagnostic testing facilities meet this rule constantly and still get caught on the purchased-component side. When a component is acquired from an outside supplier, payment is the lower of the supplier net charge, the amount billed, or the fee schedule amount, and the performing supplier name, address and NPI must be kept on file. Mobile echo vendors are exactly this arrangement. Authorization numbers also did not match the imaging performed on 7% of advanced imaging claims we audited. See radiology billing.

Ambulatory surgery centers and hospital outpatient

Facility-side echo is not billed with modifier TC at all. Hospitals report the technical side institutionally, and a facility furnishing only the technical component is not currently required to submit TC because it is paid as though TC had been billed. The error we see is a physician group inside a facility attempting a global claim, which fails against the rule barring a physician from the technical component for a hospital patient. Facility and professional claims disagreed on codes in 6% of ASC cases. See ASC billing.