BLOGS 
NEWSROOM 
CASE STUDIES 
WEBINARS 
PODCASTS 
ASSET HUB 
EVENT CALENDAR 
TL;DR
Dermatology has two revenue problems and practices worry about the wrong one.
Whether a procedure is cosmetic or medically necessary turns on the words in a clinical note and cannot be automated, which is what every training session covers, while Mohs unit counting is completely deterministic and sits on an approved CMS Recovery Audit target that almost nobody writes about. The un-encodable problem gets the attention; the encodable one gets the audit.
Every dermatology billing course I have sat through spends its time on the same question, whether a procedure is cosmetic or medically necessary, and that is a reasonable place to spend it. It is also not the thing CMS has an open audit program for.
I train practice staff on documentation and workflow, so I am not neutral about this. The cosmetic question is genuinely hard, it genuinely costs practices money, and the training is genuinely worth doing. What I did not expect, when I went looking for what actually gets audited, was to find that the answer is a piece of arithmetic almost nobody talks about.
This piece covers both. The one you cannot automate, honestly, and the one you can, which is the one carrying the audit exposure. If you want the general shape of the cycle first, our revenue cycle management services page covers it and the 13 steps of the revenue cycle is the plainer version.
Cosmetic Versus Medically Necessary: The Problem Everyone Trains On
In dermatology, the same procedure can be covered or not covered depending on why it was done. An excision, a laser treatment, a lesion removal: each can be cosmetic or medically necessary depending on clinical circumstance, and payers scrutinize the distinction closely. Cosmetic procedures are generally non-covered unless medical necessity is documented.
The determination does not live in a code. It lives in prose, in the clinical note, in whether the documentation establishes why this was treatment rather than improvement.
I want to be straightforward about what that means, because the rest of this article argues for encoding rules and this one resists it. You cannot automate this judgment. A system can prompt for it, template it, and refuse to let a claim leave without something in the field. It cannot decide whether the note in front of it establishes medical necessity, because that is a clinical and narrative determination and pretending otherwise would be selling you something.
What training does help with, and this is the practical version I give staff: the documentation has to make the distinction clearly and consistently for every service, or a medically necessary claim gets swept into the cosmetic pile by association. One loose note on one line can contaminate the encounter.
There is also a specific habit worth naming. Do not use the word “cosmetic” in a note supporting a service you are billing as medically necessary. Not as shorthand, not in passing, not describing the patient’s motivation. It is the single easiest self-inflicted denial in the specialty, and I will come back to it in a moment because it turns out to be the one part of this problem a system can actually help with.
The Problem CMS Actually Audits
That is a public, named, active audit target. Not an emerging risk, not a compliance theme. A topic with a number on it. And it is about units. Arithmetic. The most mechanically checkable thing in the entire specialty.
Set the two side by side and the asymmetry is uncomfortable. The problem that cannot be encoded absorbs most of the training budget and most of the anxiety. The problem that can be encoded completely, and could be enforced by a rule that never forgets, is the one with an audit program named after it.
I do not think that is because anyone is careless. It is because the cosmetic question feels like the hard one, and hard problems attract attention. Unit arithmetic feels like something that surely takes care of itself. It does not, and there is a CMS program that exists because it does not.
Where Mohs Units Actually Go Wrong
The Mohs code family is structured by body site and by stage, and the counting rules are specific. These come from the CMS Medicare Coverage Database billing and coding articles for Mohs Micrographic Surgery, not from a billing vendor’s cheat sheet.
- 17311 What it covers: First stage, head, neck, hands, feet, genitalia, or any site involving muscle, cartilage, bone, tendon, major nerves or vessels. Up to 5 tissue blocks
- 17313 What it covers: First stage, trunk, arms and legs. Up to 5 tissue blocks
- 17312 What it covers: Each additional stage, reported with 17311
- 17314 What it covers: Each additional stage, reported with 17313
- 17315 What it covers: Each block after the first 5 in any single stage
Three things in that table cause most of the unit errors.
- The codes count blocks, not slides. A single tissue block can yield multiple slides. Counting slides inflates the units, and inflated units on a code family with an open audit topic is exactly the pattern the audit is looking for.
- Additional-stage codes must not be reported on separate claim lines. When a Mohs procedure runs to several stages, 17312 or 17314 is not repeated line by line. The stages are totaled and entered as a single line item with the correct number of units. Splitting them across lines is a coding error even when the total is right.
- The first-stage codes include the first 5 blocks. 17315 begins at block 6. Billing 17315 for blocks that were already included is a straightforward overpayment.
There is one more scope rule worth knowing, because it catches practices that split the work. Codes 17311 through 17315 are reserved for the surgeon who removes the lesion and prepares and interprets the slides. If those functions are split across providers, this code family is not the right one, and certain biopsy, excision and pathology services are already bundled into it.
See Where Your Dermatology Coding Fails Audit Scrutiny Today
Multiple Lesions, and the Modifier That Separates Them
When Mohs is performed on genuinely separate lesions on the same date, each lesion’s first stage is reported on its own claim line with a distinct-service modifier, so the second and subsequent lesions are not bundled into the first.
Modifier 59 is the traditional way to signal that. The National Correct Coding Initiative also defines the more specific X-modifiers, XS, XE, XP and XU, and Medicare prefers those over plain 59 because they say why the services were distinct rather than just asserting that they were. XS in particular identifies a separate structure, which is precisely the situation with a second lesion.
This is deterministic. Whether two lesions are separate is a clinical fact recorded at the time of the procedure. Once recorded, which modifier applies and where it goes is a rule with no judgment in it. It belongs in claim build, not in a coder’s memory.
Modifier 25 and the Same-Day Evaluation
The other modifier that shapes dermatology revenue is 25, which identifies a significant, separately identifiable evaluation and management service performed on the same day as a procedure.
Dermatology triggers this constantly. A patient comes in about one thing, the dermatologist addresses it and also performs a procedure, and whether the evaluation was separately identifiable is a real question with a real answer that has to be documented.
Modifier 25 is a recurring, high-frequency decision, and unlike the cosmetic question it has a knowable answer at claim build if the documentation supports it.
Measure the Lesion Before You Remove It
This one is not a coding problem at all, which is why coder training never fixes it.
Excision coding depends on the size of the lesion plus the margins taken, a structure defined in the AMA CPT code set. That measurement has to happen before the excision, on the patient. Once the specimen is in formalin it has changed, and the number you needed is gone. Nobody can reconstruct it afterward, and what tends to happen instead is that a smaller, safer, defensible size gets recorded and the practice quietly undercodes.
I raise this in every training session I run and it consistently surprises people, because it does not feel like a billing issue. It is a sequencing issue in the procedure room. The fix is not education, it is a documentation template that asks for the measurement at the moment it is still available, with the field positioned before the procedure note rather than after it.
That is a workflow change costing one field in one template, and it is worth more than another hour on coding accuracy.
What Your Systems Can and Cannot Do
What a system cannot do: make the medical-necessity judgment. It cannot read a clinical note and determine whether the documentation establishes that a procedure was treatment rather than improvement. Anyone selling you that is overselling.
What a system can do, and should:
- Unit and block arithmetic. Blocks counted, not slides. 17315 starting at block 6. Additional stages totaled onto one line rather than split. This is the audited half, and it is pure computation.
- Distinct-service modifier logic for multiple lesions, with the X-modifier chosen over plain 59.
- Modifier 25 support checking at claim build, where the documentation is available to check against.
Measurement capture prompted at the right moment, which is a template change rather than a system feature.
And one small thing that punches above its weight. Flag any claim being submitted as medically necessary whose supporting note contains the word “cosmetic.” That is a single string check against a claim attribute. It costs almost nothing to build, it requires no clinical judgment, and it catches the exact self-inflicted denial I described at the top of this article. It does not solve the medical-necessity problem. It removes the dumbest version of it.
I like that check because it is honest about its scope. It is not artificial intelligence reading a note. It is a string match catching a known human habit, and those are usually the automations that actually survive contact with a real practice.
Outsource, Buy, or Build
Nearly every page ranking for this topic is published by a company selling one of these three answers. Here is the version that is not.
- Outsourcing is the right answer for a lot of dermatology practices. Under roughly 10 providers, with steady volume and nobody internally who owns billing technology, a competent partner will outperform anything you build. Outsourcing revenue cycle management covers where that line usually falls, and our list of revenue cycle management companies is a reasonable starting point for a shortlist. If the distinction between billing and the full cycle is still fuzzy, medical billing versus revenue cycle management sets it out.
- Here is the dermatology-specific caveat, and it cuts against every option. An outsourced biller cannot fix your documentation language. They can code accurately from what you give them and appeal skillfully when it denies, but the cosmetic-versus-necessary determination is made in your exam room, in your note, by your clinician. The un-encodable half of this problem stays yours regardless of who submits the claim. Any vendor conversation that implies otherwise is worth pushing back on.
- Building earns its keep when the deterministic half is where your money is going, your volume funds ongoing maintenance, and the fix has to live in a workflow you control. Given that the deterministic half is the audited half, that case is stronger in dermatology than the quiet profile of these codes suggests.
Practices choosing a platform alongside this decision should note that system selection is a separate question from billing operations, and our custom EMR and EHR development guide covers that side.
For the same argument through different specialty mechanics: cardiology revenue cycle management covers a code-set migration, gastroenterology revenue cycle management a payer-dependent decision tree, oncology revenue cycle management a mandatory attestation on the drug line, and behavioral health revenue cycle management session-based billing with carve-out payers.
Six Checks Before Your Next Audit Cycle
In the order I would run them.
- Pull your last quarter of Mohs claims and check whether units were counted by blocks or by slides. This is the audited item. Start here.
- Check whether 17312 and 17314 appear on separate claim lines anywhere. They should be totaled onto one line with a unit count.
- Confirm 17315 is only billed from block 6 onward, not for blocks already included in the first-stage code.
- Check that multi-lesion claims carry a distinct-service modifier, and whether you are using an X-modifier or defaulting to 59.
- Look at where your documentation template asks for lesion measurement. If it sits after the procedure note, move it before.
- Run one string search across last quarter’s medically-necessary claims for the word “cosmetic” in the supporting documentation. It takes minutes and the results are usually instructive.
Every one of these is diagnostic and none requires buying anything. Notice also that five of the six address the encodable half, which is deliberate. That is where the audit is
From Coding Gaps to a Repeatable Audit-Proof Process
The gap in dermatology RCM training is rarely about clinical judgment. It’s about counting. Cosmetic versus medically necessary calls will always need a physician’s narrative and clinical context, but Mohs units, lesion counts, and modifier pairing don’t. They follow rules, and rules can be checked before a claim goes out, not after CMS flags it.
The six checks above aren’t a rebuild of your coding process. They’re a diagnostic pass over the half of your revenue cycle that’s already encodable today, run in the order that catches the highest-dollar errors first. Practices that build this check into every claim cycle stop treating Mohs audits as a compliance risk and start treating them as a solved problem.
If your team is ready to see exactly where your current coding stack has these gaps, Mindbowser’s RCM accuracy assessment maps them against your last audit cycle and shows what an automated check would have caught.
It is the financial process a dermatology practice runs from scheduling through final payment. What distinguishes it from other specialties is that the practice spans medical, surgical and cosmetic services, so coverage often depends on why a procedure was performed rather than on what was performed.
The determination rests on documented clinical circumstance rather than on the procedure itself. The same excision or laser treatment can be either. Cosmetic procedures are generally non-covered unless medical necessity is documented, and payers examine this distinction closely, so the documentation has to establish it clearly and consistently for every service.
Each separate lesion’s first stage is reported on its own claim line with a distinct-service modifier so the second and subsequent lesions are not bundled into the first. Medicare prefers the specific X-modifiers over plain modifier 59 because they identify why the services were distinct.
Counting slides instead of tissue blocks. The code family counts blocks, and a single block can produce multiple slides. Two related errors are reporting additional-stage codes on separate claim lines instead of totaling them as units on one line, and billing the extra-block code for blocks already included in the first-stage code.
Modifier 25 identifies a significant, separately identifiable evaluation and management service performed on the same day as a procedure. Dermatology encounters this frequently because patients often present about one concern and receive a procedure in the same visit. The decision is supportable at claim build if the documentation distinguishes the two services.
No. That determination is clinical and narrative, and it is made in the exam room and the note. Software can prompt for it, template it and refuse to release a claim without it, and it can catch avoidable contradictions such as the word “cosmetic” appearing in a note supporting a medically necessary claim. It cannot make the judgment itself.
