Why medical claims validation may need to look beyond individual billing codes
Medical claims processing has become highly sophisticated.
Claims are checked against tariffs, benefits, membership status, authorisations, provider information and coding rules. Automated switches and adjudication platforms can process enormous volumes of transactions with remarkable speed and efficiency.
Yet there remains an important question:
What happens when every individual code on a claim is valid, but the combination of those codes is not?
This is one of the problems behind the development of QR-E.
The problem with validating codes in isolation
Traditional validation processes are designed to establish whether an individual claim item meets the required criteria.
- Is the code valid?
- Is the provider entitled to submit it?
- Does the member have the appropriate benefit?
- Does the tariff exist?
- Are the required authorisations in place?
These are all essential checks.
But healthcare claims are not always simply collections of independent billing codes. The relationship between procedures, services, equipment, facility charges and other claim items can be just as important as the validity of any individual code. A claim can therefore pass conventional validation while still containing a combination of charges that should not have appeared together i.e. every individual item can be valid. The claim can still be wrong.
When the relationship between codes matters
Consider a simple example. A provider may submit a procedure together with several additional codes. Each of those codes may be legitimate in isolation. The procedure exists. The additional services exist. The individual codes may all be correctly formatted and recognised by the claims system. But what if some of those additional services are already included within the contracted composite rate for the primary procedure? Individually, the codes may appear valid. Collectively, they may represent unbundling. This is where conventional validation can encounter a blind spot.
The question is no longer:
Is this code valid?
The more important question becomes:
Should these codes appear together on this claim?
That requires validation to examine relationships rather than simply individual data elements.
A practical example
The QR-E proof of concept illustrates this with the example of a colonoscopy claim.
In the example, a provider submitted a claim totalling R12,220 against a contracted composite rate of R7,500. The individual billing items could each appear legitimate when examined independently. However, QR-E identified three additional codes that, when considered alongside the primary procedure, represented charges already included within the applicable composite tariff. The result was an alleged excess of R4,720. The important point is not the specific procedure. It is the validation principle.
The potential problem was only visible when the claim was examined as a combination of related items.
A system looking only at whether each individual code was valid could potentially allow the claim to proceed. A system capable of examining the relationships between those codes has an opportunity to identify the issue before the claim reaches adjudication.
The cost of discovering a problem too late
Historically, medical schemes have often relied on retrospective processes to identify potentially inflated, unbundled or non-compliant claims. By that point, however, the claim may already have entered the adjudication process or been paid. The organisation then faces a more complicated set of challenges.
Potentially, this can involve:
- retrospective forensic review;
- investigation and evidence gathering;
- provider engagement;
- disputes;
- recovery processes;
- legal complexity; and
- the possibility that some value is never recovered.
The question is therefore whether certain forms of claims leakage could be addressed earlier. Rather than asking:
How do we recover money after identifying a problem?
Perhaps the better question is:
Can the problem be identified before the claim is allowed into the adjudication process?
Moving validation upstream
This is the principle behind QR-E. QR-E has been developed as a pre-adjudication validation engine intended to examine medical claims before they enter the scheme’s adjudication environment. Rather than focusing solely on whether an individual claim item is syntactically valid, QR-E examines relationships between claim items to identify potential patterns of:
- unbundling;
- incorrect combinations of codes;
- tariff non-compliance; and
- other claim relationships requiring validation.
The objective is straightforward:
Identify potentially non-compliant claim combinations before they become an expensive retrospective problem.
In practical terms, this means moving part of the validation process upstream. The intention is not to replace the scheme’s existing adjudication infrastructure. The opportunity is to add an additional layer of intelligence at the point where a claim can still be returned, corrected or prevented from progressing.
From data syntax to claim intent
Healthcare claims systems are necessarily built around structured data. Codes, formats, identifiers, authorisations and tariffs all need to be validated accurately and consistently. But structured data alone does not always explain the relationship between the items being submitted. A collection of syntactically valid codes does not necessarily represent a clinically or contractually valid claim combination. This creates an important distinction.
Data validation asks whether the information is valid. Relationship-based validation asks whether the information makes sense together.
That distinction is at the heart of the QR-E proof of concept.
The next stage: proving the concept against real-world claims
QR-E has now reached the point where the next question is not simply whether the technology can demonstrate the principle. The next question is whether it can identify meaningful patterns within real-world medical scheme claims data. That requires validation.
We are therefore interested in engaging with medical schemes and appropriate technology, claims and risk stakeholders to explore controlled validation exercises using appropriately anonymised historical claims data. The purpose would be to test QR-E against real-world claim patterns and establish:
- whether the engine identifies combinations that conventional processes may not flag;
- the nature and frequency of those patterns;
- the potential value of pre-adjudication validation; and
- whether QR-E warrants further integration and development.
No member-identifiable information would be required for such an exercise. The objective is not to ask an organisation to take a leap of faith. It is to test the proposition against real data and allow the results to speak for themselves.
A different question for claims validation
Medical claims validation has traditionally focused on an essential question:
Is this claim valid?
But as claims become more complex, there may be another question worth asking:
Are these items valid together?
That difference may appear subtle. Its implications, however, could be significant. Because sometimes the claim is not obviously wrong. Every code may be valid. Every individual item may pass. And yet, when the claim is examined as a whole, something may not belong.
The claim can be valid — and still be wrong.
About QR-E
QR-E is a proof-of-concept pre-adjudication validation engine developed to identify potentially unbundled, inflated and non-compliant medical claim combinations before they enter adjudication.
To view the QR-E proof of concept:
To discuss a controlled validation exercise using appropriately anonymised historical claims data, contact:
Ramón Fritz
enterprise@infimetrica.co.za