When a Medicare contractor or the IRS reviews a sample of your claims or transactions and multiplies the error rate across thousands of records, the resulting demand can dwarf anything the sample itself supports. A statistical sampling and extrapolation audit defense works on three fronts at once: the methodology behind the sample, the individual determinations inside it, and the procedural record the agency built along the way. Wins at any of those points can shrink a six- or seven-figure projection to a fraction of the original number, and in some cases eliminate the extrapolation entirely.
Understand What the Auditor Actually Did
Before you can challenge anything, you need to see the shape of the audit. The auditor started by defining a universe, meaning the complete population of records for the audit period. For a Medicare provider, that might be every claim submitted over three years. For a tax audit, it could be every deduction in a category. From that universe, the auditor pulled a random sample using specialized software: the Office of Inspector General’s RAT-STATS in most healthcare audits,1Office of Inspector General. RAT-STATS – Statistical Software or the LB&I Statistical Sampling Application for the IRS.2Internal Revenue Service. Statistical Sampling Auditing Techniques Often the universe is split into strata first, with high-dollar items separated from low-dollar ones so the estimate is more precise.
Once every sampled item is reviewed, the auditor calculates an error rate and multiplies. If 15 of 100 sampled claims contain overpayments averaging $200, the point estimate across a universe of 10,000 claims is $300,000. Agencies then apply a statistical cushion by demanding the lower confidence limit rather than the point estimate. CMS instructs its contractors to use the lower limit of a one-sided 90 percent confidence interval as the recovery amount.3Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual Chapter 8 – Administrative Actions and Sanctions and Statistical Sampling for Overpayment Estimation The IRS uses a 90 percent two-sided confidence level and subtracts the sampling error from the point estimate.2Internal Revenue Service. Statistical Sampling Auditing Techniques
One quiet detail worth checking in any demand letter: underpayments found in the sample must be netted against the overpayments. CMS requires contractors to record underpaid claims as negative overpayments and factor them into the extrapolation.3Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual Chapter 8 – Administrative Actions and Sanctions and Statistical Sampling for Overpayment Estimation Not every contractor does this on its own.
Demand the Documents First
Nothing else in a defense works if you cannot see how the sample was built. Before spending money on experts or drafting appeals arguments, request the technical file from the contractor or examiner. The core documents include:
- The sampling plan, describing the universe definition, stratification, sample size calculation, and selection procedure. If no written plan exists, that gap alone may support a challenge.
- The sample selection report listing every item pulled, which you cross-check against the universe definition.
- The random number seed, the starting value the software used to generate the selection. With the seed and the same software, an independent expert can rerun the draw and confirm the same items come out.1Office of Inspector General. RAT-STATS – Statistical Software
- The extrapolation calculation worksheets showing the point estimate, confidence interval, and lower confidence limit. CMS requires these to show what should have been paid on each sampled claim, whether overpaid or underpaid.3Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual Chapter 8 – Administrative Actions and Sanctions and Statistical Sampling for Overpayment Estimation
- The individual claim determinations, with the auditor’s finding and reasoning for each sampled item.
For Medicare cases, also request all correspondence between the contractor and CMS, including any approval of the sampling methodology. The Medicare Appeals Council has overturned extrapolations when a contractor failed to provide enough documentation for the provider to recreate the sampling frame, treating the gap as a due process violation.4U.S. Department of Health and Human Services. Medicare Appeals Council Decisions
Grounds That Actually Reduce the Demand
Not every extrapolation survives scrutiny. Successful challenges tend to cluster around a handful of recurring problems, and each one attacks a different piece of the projection.
Flawed Universe Definition
If the universe includes claims that should not have been reviewed at all, the multiplier is distorted from the start. Common examples include claims from a period outside the audit scope, claims already reviewed in a prior audit, or claims for services the contractor lacks authority to review. Pulling improper claims out of the universe recalculates the multiplier downward and can undermine confidence in the whole projection.
Errors in Individual Claim Determinations
Every sampled claim must be correctly evaluated. If the auditor wrongly denied claims in the sample, reversing those denials forces a re-extrapolation based on the remaining errors, which can dramatically reduce the total.4U.S. Department of Health and Human Services. Medicare Appeals Council Decisions This is often where the biggest reductions come from, because a single flipped determination in a 100-claim sample changes the projected total across the entire universe.
Insufficient Documentation of the Sampling Frame
When the record does not contain enough information to recreate the sampling frame, the provider cannot mount a meaningful challenge. Some adjudicators have treated that gap as a due process violation by the contractor and invalidated the extrapolation on that basis.
Improper Extrapolation From a Probe Review
In Medicare, a probe review is a limited initial review, not a full statistical sample. Contractors may only collect overpayments on the specific claims reviewed in a probe. Extrapolating from a probe to the broader universe is not permitted, and demands built on that misstep can be struck.4U.S. Department of Health and Human Services. Medicare Appeals Council Decisions
Sample Size Alone Is Not Enough
One argument that reliably fails: “the sample was too small.” The Medicare Program Integrity Manual says explicitly that there is no fixed minimum sample size and that challenges based solely on small size are without merit unless tied to a specific methodology flaw.3Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual Chapter 8 – Administrative Actions and Sanctions and Statistical Sampling for Overpayment Estimation Arguing that 50 claims is not enough will not work in isolation. You need to show why 50 was statistically insufficient for that particular universe, given its variability and the precision the projection claims.
IRS-Specific Grounds
The IRS holds sampling to a higher confidence threshold. Under Revenue Procedure 2011-42, a sampling estimate must be computed at the “least advantageous 95 percent one-sided confidence limit,” meaning the calculation must use whichever end of the interval produces the least benefit to the taxpayer.5Internal Revenue Service. Revenue Procedure 2011-42 If the relative precision is 10 percent or better, the taxpayer may use the point estimate directly. There is also an internal IRS rule that helps taxpayers: if the point estimate of an adjustment is positive but smaller than the sampling error, the examiner must either draw additional sample items or abandon the sampling plan entirely and propose adjustments only for the specifically identified errors.2Internal Revenue Service. Statistical Sampling Auditing Techniques
One caution on the IRS side: if the examiner determines that the taxpayer failed to maintain adequate records or did not make a reasonable effort to comply with the tax code, the adjustment can be set at the full point estimate rather than the reduced confidence-limit figure.2Internal Revenue Service. Statistical Sampling Auditing Techniques Recordkeeping isn’t cosmetic; it changes the formula.
The Statutory Limits on Extrapolation Itself
The government’s power to extrapolate is bounded. For Medicare, 42 U.S.C. § 1395ddd(f)(3) restricts when a contractor can use extrapolation at all. It is only permitted when the Secretary of Health and Human Services determines either that there is a sustained or high level of payment error, or that documented educational outreach has failed to correct the provider’s errors.6Office of the Law Revision Counsel. 42 USC 1395ddd – Medicare Integrity Program The statute also bars administrative and judicial review of the Secretary’s underlying determination that a sustained or high error level exists, which closes off one avenue but not the others. You can still attack how the sample was designed and executed, whether the individual claim determinations were correct, and whether the documentation supports the projection.
For the IRS, Revenue Procedure 2011-42 sets the criteria for whether a statistical sampling estimate qualifies as adequate substantiation. A sample that fails those criteria will be rejected.5Internal Revenue Service. Revenue Procedure 2011-42 Taxpayers can also propose their own statistical samples to support return positions, and the IRS must evaluate those proposals under the same standards.2Internal Revenue Service. Statistical Sampling Auditing Techniques
Move Quickly Through the Medicare Appeal Levels
Medicare appeals follow five levels, and each one must be exhausted before moving to the next. The clock starts with the initial determination, meaning the formal finding that an overpayment exists.7eCFR. 42 CFR 405.924 – Actions That Are Initial Determinations
Level One: Redetermination
The first level is a redetermination request filed with the contractor that issued the initial determination. The deadline is 120 days from receipt, and receipt is presumed to be five calendar days after the date on the notice. File by certified mail or the contractor’s portal so you have proof. The contractor generally has 60 days to decide.8Centers for Medicare & Medicaid Services. First Level of Appeal – Redetermination by a Medicare Contractor
Level Two: Reconsideration by a QIC
If the redetermination goes against you, the next step is reconsideration by a Qualified Independent Contractor. The deadline is 180 days from receipt of the redetermination decision, again with the five-day presumption. There is no dollar-amount threshold at this level. The QIC generally has 60 days to decide, and if it cannot, it must notify the parties and inform the appellant of the right to escalate to the next level.9Centers for Medicare & Medicaid Services. Reconsideration by a Qualified Independent Contractor
Level Three: Administrative Law Judge
The third level is a hearing before an Administrative Law Judge at the Office of Medicare Hearings and Appeals. This is the first stage where the amount in controversy matters; for 2026, the minimum threshold is $200.10Federal Register. Medicare Appeals Adjustment to the Amount in Controversy Threshold Amounts The ALJ conducts a de novo review, examining both the individual claim determinations and the statistical validity of the extrapolation. For most providers, this is the most consequential stage, because it is the first truly independent review of the sampling methodology.
Levels Four and Five: Appeals Council and Federal Court
An unfavorable ALJ decision can be appealed to the Medicare Appeals Council within 60 calendar days of receipt of the decision, with receipt presumed five days after the notice date.11eCFR. Medicare Appeals Council Review The request must identify the specific parts of the ALJ’s decision being challenged and explain why. The fifth and final level is judicial review in federal district court, which for 2026 requires a minimum amount in controversy of $1,960.10Federal Register. Medicare Appeals Adjustment to the Amount in Controversy Threshold Amounts
Recoupment Only Pauses Through Level Two
Providers often assume that filing an appeal stops collection. It does, but only for the first two levels. Federal law prohibits CMS from recouping while a redetermination or QIC reconsideration is pending. Once the second level is complete, the contractor resumes recoupment at 100 percent, regardless of whether you press on to an ALJ, the Appeals Council, or federal court. Unless the provider establishes an extended repayment schedule, the contractor withholds from future payments until the debt is fully recovered.12Centers for Medicare & Medicaid Services. Limitation on Recoupment of Overpayments
The practical effect on cash flow is significant. A provider fighting a $500,000 extrapolated overpayment through an ALJ hearing that takes a year or more will have the full amount offset against ongoing claims starting after the QIC decision. Filing quickly at each level is not a procedural formality. It directly controls how long the recoupment pause lasts and whether the practice has the cash to keep operating during the fight.
Interest and False Claims Act Stakes
Interest accrues on delinquent Medicare overpayment debts at a rate set quarterly by Treasury. As of early 2026, that rate is 11.625 percent, based on the private consumer rate.13Centers for Medicare & Medicaid Services. Notice of New Interest Rate for Medicare Overpayments and Underpayments Under 42 CFR § 405.378, the rate is the higher of the current value of funds rate (one percent for 2026) or the private consumer rate. On a six-figure extrapolated overpayment, that adds tens of thousands of dollars in the first year alone.
If the audit reveals patterns suggesting knowing misconduct rather than honest errors, the case can shift into False Claims Act territory. Under 31 U.S.C. § 3729, a person who knowingly submits a false claim is liable for three times the government’s damages plus a civil penalty for each false claim.14Office of the Law Revision Counsel. 31 USC 3729 – False Claims The per-claim penalty is adjusted annually for inflation and has been well above $10,000 per claim in recent years. When statistical sampling is used in an FCA case, the extrapolated claim count multiplied by per-claim penalties can produce staggering totals.
Courts tend to scrutinize extrapolation more closely in False Claims Act cases than in standard overpayment recoveries, particularly where the extrapolated figure drives treble damages and per-claim penalties. In straightforward overpayment cases, the government generally receives substantial deference in its use of statistical methods. Statistical sampling has not been permitted to establish the knowledge element of an FCA claim; that must be shown claim by claim. If you are facing an FCA case built on sampling, the defense needs to attack both the math and the falsity and knowledge elements of each sampled claim.
When to Hire a Statistician
Challenging an extrapolation on statistical grounds almost always requires outside expertise. A PhD statistician can evaluate whether the sampling design, execution, and projection methodology meet the applicable standards, and can serve as an expert witness if the case reaches an ALJ hearing or beyond. Hourly rates for statisticians with audit sampling experience run from roughly $200 to well over $500, depending on complexity and credentials. A forensic accountant can complement that work by reconciling the claim-level data against your billing records, catching problems like duplicate claims in the universe or miscategorized services in the sample.
The economics usually justify the spend. When an extrapolated demand runs into six or seven figures based on errors found in 100 sampled claims, flipping even a handful of those determinations can reduce the projected total by hundreds of thousands of dollars. Bring the expert in early. Waiting until the ALJ stage to discover a fundamental flaw in the sampling frame means months of recoupment that could have been avoided, and it leaves less time to build the record the ALJ actually reviews.