September 22, 2026
Healthcare Claims Data Analytics: The Healthcare Horizons Approach
By: Felicia Hall
Every healthcare claim tells part of a story. But some of the most important insights only emerge when thousands, or even millions, of claims are analyzed together.
A single claim might look perfectly reasonable on its own. Across the full claims population, however, it may stand out from a provider’s typical billing pattern, conflict with how similar claims were paid, push a member beyond a plan benefit maximum, or point to a larger configuration issue.
That ability to identify patterns across claims is fundamental to payment integrity. The Centers for Medicare & Medicaid Services (CMS) describes this type of data analysis as “a tool for identifying actual or potential claim payment errors,” including through statistical outliers and patterns within groups of claims that may signal improper billing or payment.
At Healthcare Horizons, data analytics helps us find those patterns and determine where to look — but it is only one part of the healthcare claims audit process.
We spoke with Felicia Hall, Vice President of Audits at Healthcare Horizons, about how her team turns raw claims data into focused audit investigations. Here’s how Healthcare Horizons approaches data analytics.
Step 1: We Make Sure the Data Is Complete and Accurate
Before we look for claim errors, we make sure we can trust the data we’re analyzing.
“When we receive a client’s claims data, we’re reviewing to make sure their annual spend matches the received data from the TPA and all groups and plans are accounted for,” Hall explains.
Our team also reviews field accuracy and confirms that the data matches the specifications in the data dictionary and file layout provided by the third-party administrator (TPA).
These initial checks establish the foundation for the audit. If a group or plan is missing from the dataset, for example, we’re not looking at the client’s full claims experience. Similarly, if fields are missing or incorrectly formatted, they can affect the checks we’ll perform later.
Once we’ve validated the dataset, we can begin looking for what it can tell us.
Step 2: We Analyze the Full Claims Population
Once the data is prepared, we run it through more than 75 standard checks to identify potential errors and anomalies. We look for duplicate claims, missed coordination of benefits, ineligible members, medical edits, incorrect discounts, non-participating provider pricing, unusual units or payments, and other areas of concern.
But the analysis isn’t one-size-fits-all.
We also review each client’s plan documents and build custom checks around how that plan should operate, including its deductibles, copays and coinsurance, out-of-pocket maximums, benefit limits, and exclusions. According to Hall, these client-specific checks can roughly double the number performed during an audit.
Together, these checks form part of our healthcare claims audit methodology, allowing us to test claims against both standard payment rules and the terms of the client’s own health plan.
Step 3: We Identify Patterns and Outliers
Analyzing 100% of a client’s healthcare claims gives us something individual claims can’t: context.
“If a certain provider is consistently being paid at a certain percentage of billed charges and we find a handful of claims allowed at a different rate, we’re able to flag those as potential errors,” Hall explains.
The same principle applies elsewhere. If a member’s claims consistently show coordination with another primary insurer, for example, claims without that coordination stand out. If a plan limits physical therapy to 30 visits per calendar year, we can review all of a member’s physical therapy visits to see whether the limit was exceeded.
We also look at utilization over time, including members with unusually high numbers of ER visits and providers billing more procedures in a day, month, or year than reasonably possible. Unusually high units or charges can expose manual keying errors that might otherwise blend into the claims population.
“Reviewing the entire dataset helps us find those claims that are outside the norm,” Hall says.
Those outliers give our auditors a more focused place to start.
Step 4: Our Auditors Investigate the Results
Analytics can flag a claim as unusual, but unusual doesn’t necessarily mean incorrect. That’s where our auditors come in.
Once we’ve narrowed the results from our standard and custom checks, our auditors manually review the potential errors and focus on those most likely to warrant further investigation.
That review can uncover incorrect patient responsibility, duplicate billing, missed coordination of benefits, and surgical bills split in ways that result in full payment for multiple services. It also helps us distinguish payment errors from legitimate exceptions in the data.
As CMS explains in its Medicare Program Integrity Manual, analytical methods identify potential errors that require verification before further action is taken.
At Healthcare Horizons, analytics tells us where to look. Our auditors determine what the data means.
Step 5: We Look for the Bigger Issue Behind the Error
Sometimes an error is isolated. Other times, it points to a problem affecting many claims.
Some errors stem from manual claims processing. With numerous edits, pricing rules, benefit limits, exclusions, authorizations, and other requirements to consider, a missed edit or simple keying mistake can result in an incorrect payment.
Other errors originate in the system itself. Hall points to incorrectly configured patient responsibility as an example. If a copay, deductible, or coinsurance rule isn’t loaded correctly for a particular service, the difference on any one claim may be small. Repeated across the claims population, however, those errors can add up to a significant amount.
The same is true of other system configuration issues. A rule or edit that isn’t loaded correctly can produce the same error again and again until the pattern is identified.
That’s why we don’t stop at asking whether an individual claim was paid correctly. We also ask whether the finding points to a larger issue in how the plan is being administered.
From Claims Data to Greater Confidence in Plan Performance
The goal of data analytics isn’t to find errors for the sake of finding errors. It’s to give clients a clearer picture of how their health plan is being administered.
That may mean uncovering a missed edit, benefit configuration issue, or payment pattern that needs attention. It may also confirm that the TPA is processing claims accurately and administering the plan as intended.
“The TPA is a great partner in administering your plan,” Hall says. “We come in to verify that everything is running smoothly and there are no areas of concern.”
By combining data analytics with experienced human review, Healthcare Horizons helps clients understand not just what their plan paid, but whether it paid correctly.
Because knowing what you spent is only part of the story. Knowing it was paid correctly is what builds confidence in your plan.
Want greater confidence in your healthcare claims? Contact our team today
About Healthcare Horizons™
Healthcare Horizons is a healthcare audit and advisory firm dedicated to protecting the financial integrity of employee benefit plans. As a trusted partner to employers, brokers, and payers, we conduct independent healthcare claims audits to identify overpayments, uncover systemic errors, and confirm that plan administration aligns with contractual terms.
Our investigative, root-cause methodology reviews 100% of claims at the transaction level, combining advanced algorithms with deep human expertise to detect discrepancies that automated systems often miss. As a trusted partner and strategic extension of employers, we translate findings into practical recommendations that help organizations recover funds, prevent recurring issues, and strengthen plan performance.
