A screening colonoscopy becomes diagnostic the moment the physician intervenes, not when they see something. Append modifier PT to the diagnostic CPT code, never the screening G code, and sequence a screening diagnosis first. In our audit of 9,800 GI claims, 14% of converted screenings carried neither PT nor modifier 33.
Most practices treat modifier PT as a coding detail, and it is really a patient-communication failure wearing a coding costume. In our audit of 9,800 GI claims, 14% of screening colonoscopies that converted to diagnostic were billed without the PT or 33 modifier, and every one of those sent a patient a larger bill than the rules allowed. We think the fix belongs at check-in rather than at coding: a practice that tells the patient beforehand what a polypectomy will cost them stops most of the calls, the complaints and the write-offs that follow.
Methodology:Figures credited to Luxen come from four datasets. Luxen client data covers 38 client practices, January 2024 to June 2026. Luxen billing reviews covers 410 practice billing reviews, January 2025 to June 2026. Luxen claim audit covers 61,400 claims audited, January 2025 to June 2026, of which 9,800 are gastroenterology claims. The Luxen Practice Manager Survey 2026 covers 286 practice managers, March 2026. Public figures are cited to the publishing body with the year the rule takes effect, not the year it was published.
The conversion happens in the endoscopy suite, not the billing office, and it takes about four seconds. The scope is in, the physician sees something, an instrument goes down the channel. From that moment it is a different claim from the one the patient was scheduled for, and the patient does not know it yet.
A screening colonoscopy becomes diagnostic the moment the physician intervenes, not when they see something. Photographing a polyp is still a screening. Removing, biopsying or ablating it is not.
CMS states the trigger in Article A55069: if during a colonoscopy a pathology is encountered that necessitates an intervention which converts the screening colonoscopy to a diagnostic or therapeutic colonoscopy, the appropriate CPT code which includes the PT modifier must be submitted with an appropriate diagnosis to justify the procedure.
Two things follow, and most teams get the second one wrong. First, intent at the start of the case makes it a screening. A patient who arrives with rectal bleeding was never on a screening pathway, so nothing converts and PT does not belong on the claim at all. Second, the converted claim does not carry both codes: the diagnostic CPT code replaces the screening G code. You do not bill G0121 and 45385 for the same session.
Surveillance is the genuinely unsettled one, and every page on this topic ducks it. Neither CPT nor CMS defines it as a category. Medicare reads a patient with a personal history of adenomatous polyps as high risk under 42 CFR 410.37 and pays G0105 on the shorter interval. Commercial plans often read the same patient as diagnostic and apply cost sharing. There is no modifier that settles the argument. Document the personal history, bill the screening pathway, and expect the appeal. The cash-flow effect of that gap sits in our gastroenterology revenue cycle breakdown.
Modifier PT is a HCPCS Level II modifier. The CMS descriptor reads: colorectal cancer screening test; converted to diagnostic test or other procedure. It tells Medicare this claim began as a covered screening even though the code sitting on it is a diagnostic one.
It goes on the diagnostic or therapeutic procedure code, never the screening G code. Noridian sets the scope as HCPCS codes 10000 to 69999 plus G0500, and adds that it should also be attached to the associated anesthesia codes. That anesthesia line is the one most often missed. Anesthesia claims for GI endoscopy were denied 2.4 times as often as the procedure claims in our audit, and a converted screening carrying PT on the scope but not 00811 is one reason.
Watch the descriptor drift. Noridian words the same modifier as a colorectal cancer screening test which led to a diagnostic procedure. The meaning is identical and the CMS wording governs, but a coder comparing the two pages and concluding there are two modifiers is a real failure mode, and it is why certified coders build these rules once rather than per claim.
PT is a Medicare construct and the coinsurance relief behind it is Medicare law, so it works properly only there. Commercial payers accepted modifier PT without a denial on 41% of converted screening claims we reviewed, so it is tolerated more often than the textbooks suggest. Tolerance is not policy: a plan that ignores PT today can start rejecting it on a January edit release, and nothing in the ACA protection below depends on the modifier being there.
CMS sets out three steps in A55069. There are three more the article does not mention, and they are where the money goes.
A55069 supports Z12.11, Z80.0 and Z15.060. That third one matters because it is new: A55069 is on revision R11, effective 1 October 2025, and Z15.060, genetic susceptibility to colorectal cancer, was added in that revision. Every competing page on this topic still cites the 2023 version of the article, so none of them carry it. If your scrubber was built before October 2025, it does not have it either.
The deductible is waived. The coinsurance is not, and the percentage depends on the year of the date of service.
Coinsurance on a converted screening was the standard 20% in 2022, falls to 15% for 2023 through 2026, 10% for 2027 through 2029, and zero on or after 1 January 2030, with the deductible waived in every band.
The trap is 2022, and nobody flags it. MM12656 carries an effective date of 1 January 2022 and an implementation date of 1 January 2023. Read quickly, that looks like the phase-down started in 2022. It did not: the ordinary 20% applied through calendar year 2022, and the first reduced rate applies to dates of service on or after 1 January 2023. A practice reworking old 2022 claims at 15% is rebilling to the wrong number.
Take a GI practice where the allowed amount on a snare polypectomy is $415 in its locality. The same converted screening, billed correctly with PT, bills the patient $83 on a 2022 date of service, $62.25 today, $41.50 in 2028 and nothing from 2030. At 20 converted screenings a month, that is $1,245 a month in balances nobody warned the patient about at check-in. The front desk takes that call, not the coder: 1 in 9 patient balance calls was about a preventive visit billed with a cost share. Telling the patient beforehand is cheaper than explaining afterwards, which is why we build it into patient billing setup.
The regulation is not written in years. It is written in months, and that is why frequency denials keep landing on claims the practice was sure were clean. 42 CFR 410.37 requires that at least 119 months have passed for an average-risk screening colonoscopy, and at least 23 months for a high-risk one. Screening flexible sigmoidoscopy is 47 months, and a screening fecal-occult blood test is 11 months.
119, not 120. A claim at month 118 denies and one at month 120 pays. 1 in 5 frequency denials on screening colonoscopy claims we appealed had a prior screening 118 or 119 months earlier, which is a scheduling problem wearing a billing costume. The crossover rule catches people too: an average-risk screening colonoscopy after a prior screening flexible sigmoidoscopy is payable only after at least 47 months, not 119, and almost no published guidance states it. Pulling the prior screening date during eligibility verification costs nothing and prevents the whole category.
The USPSTF final recommendation is dated 18 May 2021: grade A for adults 50 to 75, grade B for adults 45 to 49. That is where the age 45 comes from, and it is a 2021 USPSTF recommendation, not a CMS change in 2023. The regulation is narrower still. 42 CFR 410.37 bars payment for a screening fecal-occult blood test or a screening flexible sigmoidoscopy performed for an individual under age 45, and states no minimum age at all in the screening colonoscopy provisions. MAC pages summarise the benefit as starting at 45, which is the sensible operating rule, but an appeal quotes the regulation.
This is a daily decision in 2026 and no page in the results lays it out. Since 1 January 2023, modifier KX on a screening colonoscopy code means the colonoscopy is a follow-on to a positive non-invasive stool-based test. MM13017 is explicit that cost sharing does not apply to the stool-based test or the follow-on colonoscopy, because both are specified preventive screening services. From 1 January 2025, 42 CFR 410.37 folds that follow-on colonoscopy into the definition of a colorectal cancer screening test outright, including after a positive blood-based biomarker test.
The decision runs in two steps. Positive stool test, colonoscopy performed, nothing removed: bill the screening code with KX. Positive stool test, colonoscopy performed, polyp removed: the case has converted, the screening code is gone, and PT goes on the therapeutic code. One says why the screening happened, the other says what it turned into.
One warning, because a widely read competing page gets this wrong: the KX in colorectal screening is not the KX from the therapy threshold. Same two letters, unrelated rule.
Non-Medicare plans do not run on modifier PT. They run on the ACA preventive services rule, and the answer is more settled than the vendor pages admit.
The tri-agency FAQ Part XII, dated 20 February 2013, answers it directly: polyp removal is an integral part of a colonoscopy, so a plan or issuer may not impose cost sharing for a polyp removal during a colonoscopy performed as a screening procedure. FAQ Part 54, dated 28 July 2022, restates it and extends it explicitly to the anesthesia provided in connection with the screening. FAQ Part 51, dated 10 January 2022, adds that a follow-up colonoscopy after a positive stool-based test must be covered without cost sharing, for plan years beginning on or after 31 May 2022.
The limit sits in the same answer, and it is the sentence plans quote back at you: a plan may impose cost sharing for a treatment that is not a recommended preventive service, even if the treatment results from one. Polypectomy is inside the screening. Treating a lesion found later is not. That distinction is the whole appeal, and one citing FAQ Part XII, Q5 by name lands differently from one asserting that the ACA covers screenings. Appeals we filed were overturned 68% of the time. Our denials and AR recovery team works these as a standing monthly batch, and the broader mechanics are in our guide to denial reasons and how to appeal them.
14% of screening colonoscopies that converted to diagnostic were billed without the PT or 33 modifier, across 9,800 GI claims in our audit. That is the headline number, and the rest of the failures cluster behind it.
Missing PT or 33 accounts for 29% of converted screening denials, the screening diagnosis not sequenced first for 21%, an interval that was too recent for 18%, the modifier sitting on the wrong claim line for 14%, and a bare anesthesia line for 10%.
Keep it in house if someone owns the endoscopy code set and reads MAC bulletins, because these rules change on bulletin cycles rather than annually and the work is not hard. Move it outside when nobody does. Fully loaded in-house billing cost 7.9% of collections for practices under $2M, against 3% to 6% for outsourced billing, and the decision usually turns on coverage rather than price. Our guide to choosing a medical billing company covers what to ask, and the wider picture is in revenue cycle management.
Want to know how this applies to your practice? We will review your AR and denials, free, in 30 minutes.
Book the reviewThe rule most pages give you, that modifier 33 is commercial and modifier PT is Medicare, is close enough to be useful and wrong often enough to cause denials. Medicare does accept modifier 33, on a short and specific list of codes, and knowing when to use modifier 33 for colonoscopy work is mostly about knowing that list.
| Question | Modifier 33 | Modifier PT |
|---|---|---|
| What is it | CPT Level I modifier for a preventive service | HCPCS Level II modifier for a colorectal screening that converted |
| What it asserts | This service is preventive under a USPSTF A or B rating or a preventive mandate | This diagnostic code started life as a covered screening |
| Primary audience | Commercial and ACA-governed plans | Medicare |
| Does Medicare take it | Yes, narrowly: G0500 and 99153 moderate sedation in support of a screening colonoscopy, and G0136, 99497 and 99498 same date as G0438 or G0439 | Yes, this is its home |
| Which line | The preventive service line | The diagnostic or therapeutic line, plus the anesthesia line |
| Both on one line | No. Noridian states modifier 33 and modifier PT should not be submitted on the same claim line | |
| Effect on the patient | Asserts a status the plan then adjudicates; it does not by itself guarantee zero cost sharing | Waives the Part B deductible and applies the reduced coinsurance for that year |
| Where it fails | Grandfathered plans and out-of-network care sit outside the ACA rule | Cases that were diagnostic from the start, where nothing converted |
The practical test is what the case was when it started, not who the payer is. If it began as a screening and an intervention happened, it converted, and the payer decides which modifier records that.
This is the home case, and the leak is concentrated. 14% of converted screening colonoscopies carried neither PT nor modifier 33 in our audit of 9,800 GI claims, and multiple endoscopy base-code ranking errors underpaid 9% of multi-procedure sessions on top of that. A GI practice running 80 screenings a month with a 30% conversion rate is making this decision 24 times a month. Build the interval check and the modifier rule into the scrubber once rather than trusting the coder to remember which of the two applies. The wider specialty picture is on our gastroenterology billing page.
The ASC carries the facility claim and the physician the professional claim, and the two have to agree the case converted. Facility and professional claims disagreed on codes in 6% of ASC cases we audited, and a converted screening is a likely place for it: the physician bills 45385 with PT, the ASC bills the screening code, and the patient gets two statements with two different cost shares. Reconcile the operative report against both claims before either goes out.
Anesthesia is where the deductible quietly reappears. Noridian instructs that PT be attached to the associated anesthesia codes, and 00811 sits inside the protected 10000 to 69999 range alongside G0500 for moderate sedation. When the anesthesia group is a separate entity, nobody owns telling them the case converted, so build a standing notification rather than relying on the operative note reaching them.
Once a polyp is removed it goes to pathology, and that specimen generates its own claim under a different rule. The pathology charge is diagnostic on its own terms and is not protected by the conversion, which surprises patients told the whole thing was preventive. Timing makes it worse: pathology charges were billed a median of 11 days after the procedure in our billing reviews, so the surprise arrives as a second statement weeks later.
Primary care bills none of these codes and absorbs all of the fallout. The referring practice ordered the screening, told the patient it was free, and takes the call when it was not. It also holds the two records that decide the claim: the prior screening date governing the 119-month or 23-month interval, and the family history that makes the patient high risk under 42 CFR 410.37 rather than average risk. Sending that history with the referral changes which code the GI practice bills.
No. When a screening colonoscopy converts, the diagnostic or therapeutic CPT code replaces the screening G code rather than joining it. You bill 45385 with modifier PT and sequence a screening diagnosis first. Billing both codes for one session is a duplicate that denies, and it is one of the most common errors we see on converted screening claims.
Yes. Noridian instructs that modifier PT be attached to the associated anesthesia codes as well as the procedure code, and 00811 and G0500 both sit inside the protected HCPCS range of 10000 to 69999. Leaving the anesthesia line bare charges the patient a deductible that the rest of the claim waived, and it is the single most commonly missed line.
The screening diagnosis goes first and the finding goes second. CMS Article A55069 supports Z12.11, Z80.0 and Z15.060, the last of which was added in revision R11 effective 1 October 2025. Sequencing the polyp ahead of the screening code tells the payer the case was diagnostic from the outset, which removes the patient protection you were trying to claim.
It waives the deductible but not the coinsurance. Medicare policy waives the Part B deductible for surgical procedures furnished on the same date and in the same encounter as a colonoscopy initiated as a colorectal cancer screening. Coinsurance is separate and phases down by year: 15% for 2023 through 2026, 10% for 2027 through 2029, and zero from 2030.
No. Noridian states directly that modifier 33 and modifier PT should not be submitted on the same claim line. They make overlapping assertions, so a payer receiving both reads a contradiction. Medicare does accept modifier 33 in narrow situations, including G0500 and 99153 moderate sedation furnished in support of a screening colonoscopy, but not stacked with PT.
The claim usually pays the practice the same amount and overcharges the patient. Without PT, Medicare treats the procedure as an ordinary diagnostic colonoscopy, applies the full deductible and the standard coinsurance, and nothing on the remittance flags the error. That is why this leak runs for months undetected: the practice sees payment, and only the patient sees the problem.
A free 30 minute review of your AR ageing and denial reasons. We tell you what is recoverable and what it would take. No deck, no commitment, no fee.
Book a free billing review