Payment Integrity
Commercial Price Intelligence for Healthcare Payment Integrity
Add negotiated-rate benchmarks and intelligent comparable selection to claim screening, overpayment research, and FWA investigation.
In short
Price transparency adds an independent commercial-pricing signal to healthcare payment integrity systems. Gigasheet returns direct negotiated-rate benchmarks for claims where the code, provider, payer, and market are known, and uses AI-assisted comparable selection to reason across provider, payer, service, geography, sample size, and customer rules when the best comparison is not an exact match. The output is market evidence that helps prioritize review, not a determination of fraud or overpayment.
What is healthcare payment integrity?
Payment integrity is the discipline of making sure a medical claim is paid correctly: to the right provider, for a covered member, at the contracted or otherwise defensible amount, once. It runs on both sides of adjudication.
| Stage | What it does | Typical checks |
|---|---|---|
| Pre-payment | Stops incorrect payments before money moves | Duplicate detection, coding edits, eligibility, prior authorization, contract pricing, medical necessity |
| Post-payment | Finds and recovers payments that were already made incorrectly | Data mining, clinical audit, coordination of benefits, subrogation, overpayment recovery |
Most of that work asks whether the claim was processed correctly against the rules that govern it. A different question sits alongside it: whether the amount paid is reasonable relative to what comparable payers and providers agree to in the commercial market. Coding accuracy and price reasonableness are separate findings, and a claim can pass one while failing the other.
What does commercial price intelligence add?
Claim edits are built to answer known questions from data the plan already holds. Commercial rate intelligence answers a question that data cannot: how the payment compares with the market.
| Signal | Answered by | Data it needs |
|---|---|---|
| Duplicate submission | Claim edits | Claim history |
| Coding and bundling errors | Claim edits, NCCI logic | Claim lines, code sets |
| Eligibility and coverage | Claim edits | Enrollment data |
| Contract pricing exceptions | Contract engine | Contracted fee schedule |
| Medical necessity | Clinical review | Clinical records, policy |
| Price reasonableness versus the commercial market | Commercial rate benchmarks | Negotiated rates from Transparency in Coverage and hospital MRFs |
| Price position versus Medicare | Medicare reference rates | Geographically adjusted Medicare rates |
The last two rows are what Gigasheet supplies. They do not replace the rows above them. They give analysts and models an external reference point that internal claim data cannot provide on its own.
Payment integrity workflows Gigasheet can support
Benchmark high-dollar claims against relevant commercial rates
Compare the allowed amount on inpatient, outpatient, and professional claims above a dollar threshold with the distribution of negotiated rates for the same service, comparable providers, and comparable payers in the same market.
Surface provider and facility pricing anomalies
Identify providers whose negotiated or paid rates sit consistently above the commercial distribution for their peer group, so review effort concentrates where the pricing signal is strongest.
Compare allowed amounts with Medicare and commercial reference points
Express each payment as a position within the commercial rate distribution and as a multiple of geographically adjusted Medicare, two lenses that often disagree and are more informative together.
Assess payer and network reimbursement patterns
Evaluate how a payer, plan, or network reimburses a service line relative to competing commercial arrangements in the same geography.
Generate pricing features for existing FWA models
Return rate percentiles, distribution statistics, and Medicare multiples as structured features that feed an existing risk score, rather than replacing the model.
Prioritize claims for deeper investigation
Rank flagged claims by the strength of the pricing signal and the confidence of the comparison set, so analysts open the most defensible cases first.
Use deterministic lookups for scale, reasoning for exceptions
Payment integrity systems process millions of claims. Almost all of them can be benchmarked with a direct query. A small share cannot, and those are frequently the claims that matter most.
| Tier 1: Direct API lookup | Tier 2: AI-assisted comparable selection | |
|---|---|---|
| Trigger | Code, provider, payer, and geography are all known and matched | Ambiguous market, thin exact matches, unusual provider, no payer match, or a customer comparison policy applies |
| Method | Deterministic query against negotiated-rate data | Reasoning across service, provider, payer, geography, sample sufficiency, and customer-defined rules |
| Latency and volume | Batch or real-time, millions of claims | Seconds per claim, exception volume |
| Output | Matching rates, percentiles, aggregates | Selected comparison set, benchmark statistics, and the underlying rate records on request |
| Where it lands | Pre-pay edit engine, risk score, data warehouse | Analyst queue, investigation workspace |
The two tiers share the same underlying data. Tier 1 answers what the rate is for this exact combination. Tier 2 answers what the most relevant comparison is, given what is known and what the customer's policy allows.
Tell Gigasheet what you are trying to determine, not only the exact row you want. The AI-assisted endpoint evaluates available rate data, applies your comparison rules, and selects the most relevant commercial comparables available.
Example: investigating an unusual inpatient payment
Illustrative. Values are synthetic and do not represent any real provider, payer, or customer.
Claim inputs
| Attribute | Value |
|---|---|
| Service | MS-DRG 470, major joint replacement of lower extremity without MCC |
| Facility | Community hospital, 180 beds, non-teaching |
| Payer | Regional commercial PPO |
| Market | Mid-size metro, Southeast |
| Allowed amount | $58,400 |
Step 1: Exact-match attempt
The direct lookup finds the same payer and facility with a negotiated rate for MS-DRG 470 at $41,200. The allowed amount is 42% above the contracted rate for this exact combination. That alone is a contract exception worth routing to review.
Step 2: Is the contracted rate itself reasonable?
The analyst wants to know whether $41,200 is high for the market before deciding how hard to push. The exact facility-and-payer pair is one observation, not a distribution. The AI-assisted endpoint is invoked with the customer's comparison policy.
Step 3: Comparable selection
Following the customer's approved rules, the endpoint evaluates:
- Same MS-DRG, no service widening permitted for inpatient surgical DRGs
- Facilities: non-teaching community hospitals in the 100 to 300 bed range within the same metro, then the state
- Payers: the same payer first, then other commercial PPOs, excluding Medicare Advantage and Medicaid managed care
- Sample threshold: at least 12 negotiated rates before reporting a distribution
- Result: 19 comparable rates selected from 7 facilities and 3 commercial PPOs within the metro; no geographic widening needed
Step 4: Benchmark output
| Statistic | Value |
|---|---|
| 25th percentile | $34,900 |
| Median | $38,600 |
| 75th percentile | $43,100 |
| Contracted rate ($41,200) | 68th percentile |
| Allowed amount ($58,400) | Above the maximum observed comparable |
| Geographically adjusted Medicare | $13,800 |
| Allowed amount as multiple of Medicare | 4.2x |
| Median comparable as multiple of Medicare | 2.8x |
Analyst takeaway
Two findings, not one. The payment exceeds the contracted rate by 42%, which is a contract exception. Separately, the contracted rate itself sits in the upper third of comparable commercial rates for this market. The first finding supports recovery. The second informs the next contract negotiation. Neither finding establishes that the claim was coded incorrectly, that the case lacked medical necessity, or that the payment was improper under the contract's outlier or carve-out terms, all of which require review of the claim record and the agreement.
How should pricing anomalies be used in FWA analysis?
A pricing anomaly is a signal, not a finding. Fraud, waste, and abuse are determinations that depend on intent, pattern, documentation, and policy, none of which are visible in a negotiated-rate dataset. In practice, commercial rate intelligence supports FWA programs in three ways:
- As a feature: rate percentile, Medicare multiple, and comparison-set confidence become inputs to a risk model alongside coding, utilization, and provider-history features.
- As a prioritization key: among claims already flagged for other reasons, the strength of the pricing signal helps decide which cases an investigator opens first.
- As supporting evidence: when a case proceeds, the selected comparison set and the underlying rate records document how the market benchmark was constructed.
What it does not do is establish that a provider billed improperly. A high rate can reflect a legitimately negotiated contract, a specialty designation, or an outlier provision. The benchmark tells the investigator where to look. It does not tell them what they will find.
Integration pattern
| Stage | System | What happens |
|---|---|---|
| 1 | Claims platform or data warehouse | Claims are staged with code, provider NPI, payer, and geography |
| 2 | Gigasheet direct API | Every claim receives a rate benchmark, percentile, and Medicare multiple in batch or real time |
| 3 | Routing rule | Claims with a strong pricing signal, thin comparison sets, or a matching customer policy are routed to Tier 2 |
| 4 | Gigasheet AI-assisted endpoint | Exception claims receive a reasoned comparison set with benchmark statistics and, on request, the rate records used |
| 5 | Risk score, analyst queue, or investigation workspace | Outputs land as features, priorities, or case evidence in the tools the team already uses |
Teams that need the full rate dataset in their own environment for model training or retrospective analysis can license it through Bulk Data instead of, or alongside, the API.
Frequently Asked Questions
What is healthcare payment integrity?
Payment integrity is the set of pre-payment and post-payment processes that ensure medical claims are paid to the right provider, for an eligible member, at the correct amount, once. Pre-payment work prevents incorrect payments through edits, eligibility checks, and contract pricing. Post-payment work identifies and recovers payments already made in error through data mining, audit, and coordination of benefits.
Can price transparency data identify claim overpayments?
It can provide independent benchmark evidence that a payment is high relative to commercial negotiated rates or Medicare. Whether that payment is an overpayment depends on the governing contract, coding accuracy, eligibility, and adjudication rules, which price transparency data does not contain. The benchmark supports the case; it does not close it.
How can negotiated rates support FWA analysis?
As pricing features and anomaly signals. Rate percentiles, Medicare multiples, and comparison-set confidence can feed an existing risk model, prioritize which flagged claims an investigator opens first, and document how a market benchmark was built when a case proceeds.
What is the difference between a claim edit and a commercial rate benchmark?
A claim edit tests a claim against rules and data the plan already holds: duplicates, coding logic, eligibility, contract terms. A commercial rate benchmark compares the payment with rates negotiated by other payers and providers in the market. Edits establish whether the claim was processed correctly. Benchmarks establish whether the price is reasonable. Both are needed.
Can Gigasheet benchmark a high-dollar claim?
Yes. When the code, provider, payer, and market are known, the direct API returns the negotiated rate and its position in the commercial distribution. When an exact match is thin or unavailable, the AI-assisted endpoint selects comparables across facility type, payer set, geography, and service, following the customer's approved rules, and returns the comparison set with its statistics.
What if there is no exact provider, payer, or geography match?
The AI-assisted endpoint evaluates alternative comparison paths: comparable facilities, a broader set of commercial payers, an expanded geography, or a customer-approved service family. It applies them in the order the customer permits and only reports a benchmark once the comparison set meets the minimum sample threshold.
Can customer-specific payment integrity rules be incorporated?
Yes. Comparison ordering, weighting, exclusions, sample thresholds, and fallback logic are configurable. A customer can, for example, prohibit service widening for surgical DRGs, exclude Medicare Advantage rates from commercial comparisons, or require a minimum number of facilities before a distribution is reported.
Does Gigasheet determine whether a claim is fraudulent?
No. Gigasheet supplies pricing intelligence and the evidence behind it. Fraud, waste, and abuse determinations depend on intent, documentation, pattern, and policy, and remain with the payment integrity team's investigators and clinical reviewers.
