We already have a claim scrubber.
That is a fair objection.
Modern claim scrubbers can do much more than check whether required fields are populated. Commercial platforms advertise payer-specific edits, configurable rules, custom claim logic, and AI-assisted rule creation. Waystar, for example, reports 2.5 million continuously updated edits and configurable payer rules. Experian describes both general and payer-specific edits in its Claim Scrubber. (6)(7)
The useful question is:
What did the scrubber have enough information to evaluate?
That distinction matters when a reimbursement decision depends on more than the fields carried on the claim.
Most difficult claim questions involve three things
A reviewer usually needs to establish:
- What is being billed?
- What rule applies?
- What does the available evidence support?
Claim edits can address a meaningful part of that work.
Some can evaluate code relationships, units, required fields, payer-specific rules, and other conditions that can be represented in structured data.
But passing the edits applied to a claim does not necessarily establish the third point: whether the underlying record supports the billing decision.
CMS's own claims process makes that distinction visible.
Medicare does not treat every edit as the same question
The National Correct Coding Initiative is a useful example, with an important caveat: NCCI is not a proxy for every commercial claim scrubber.
It is useful because CMS is explicit about what its edits do.
NCCI Procedure-to-Procedure edits address code combinations that generally should not be reported together. Medically Unlikely Edits address units of service. CMS also states that NCCI edits are correct-coding edits, not medical-necessity edits, and that the NCCI program does not perform medical review. (1)(2)
CMS describes its electronic claims process in layers as well.
A Medicare claim first passes basic HIPAA edits. It then passes implementation-guide edits. Only after those checks does Medicare evaluate the claim against its coverage and payment-policy requirements. (3)
Passing one layer tells you something useful.
It does not answer every question that can determine whether the claim is supported.
The claim and the record answer different questions
A claim tells you what is being billed.
The record can establish why that billing choice is supported.
CMS requires sufficient documentation to verify that Medicare services were performed, were reasonable and necessary, and supported the level of care billed. CMS also treats documentation as insufficient when a reviewer cannot determine whether the service was provided as billed, was medically necessary, or satisfied a required condition of payment. (4)
That means a claim can pass the edits applied to it while an important reimbursement question remains unresolved.
The codes, units, and modifier may all pass the applicable checks.
The record still has to support what happened in the encounter.
Modifier 25 makes the difference concrete
Consider an E/M service performed on the same day as a minor procedure.
Under Medicare NCCI guidance, modifier 25 may be appropriate when the E/M service is significant and separately identifiable from the other service performed that day. CMS also says modifiers should not be appended simply to bypass an NCCI edit when the clinical circumstances do not justify their use. (5)
A claim-level system can see the codes.
It can see that a modifier is present or absent.
It may know that the code relationship permits a modifier under the right circumstances.
But there is another question:
Does the record show that those circumstances actually existed?
For modifier 25, that can mean determining whether the documented E/M work was separately reportable from the work already associated with the procedure.
That is an evidence question.
The fact that a modifier can be used does not, by itself, establish that it should be used on this encounter.
This is a better way to evaluate a claim scrubber
The product label matters less than the information the system can evaluate.
Ask what the system can actually see. Can it identify the payer requirement that applied on that date of service, evaluate the documentation needed to support it, and show the reviewer what the evidence does and does not establish?
If the answer to all of those is yes, the system is already doing more than basic claim scrubbing, regardless of what the vendor calls it.
If the answer stops at the claim or the edit, someone may still have to reconstruct the rest.
That reconstruction is where difficult claims can pull experienced people back into the workflow.
They have to find the rule, determine whether it applies, inspect the available evidence, and decide what the evidence supports.
The goal is not another edit
Adding more alerts is not automatically useful.
The reviewer still needs to understand why something deserves attention.
For difficult reimbursement questions, that means bringing the claim, the applicable rule, and the available evidence together before asking someone to act.
Missing or conflicting evidence should stay visible.
Where judgment remains, a qualified person should make the decision.
That is a different objective from simply producing another pass or fail result.
Where Orchid fits
Orchid is being built around that evidence layer.
Orchid Review starts after adjudication. It helps teams understand where recurring reimbursement issues appear, what rules and evidence apply, and which questions deserve attention.
The same principle carries upstream with Orchid Prevent: bring the applicable rule and available evidence to the claim before submission, while the team can still decide whether anything should change.
Orchid is designed to work alongside the existing claim and submission workflow rather than replace it.
A capable claim scrubber remains part of that workflow.
The remaining problem begins when a billing decision depends on a payer rule, supporting documentation, or uncertainty that the existing edit did not resolve.
When the claim needs more than an edit, the reviewer should not have to reconstruct the evidence from scratch.