How to Challenge a Statistical Sample and Extrapolation in a Medicare Overpayment Demand

An extrapolated Medicare overpayment demand can be many times larger than the denied amount in the sample. Because the stakes are high, providers often ask whether they can challenge the statistical sample and extrapolation, and how.

The answer is yes, but with important limits. Providers can generally challenge two things: the individual determinations on the sampled claims and the validity of the statistical sampling and extrapolation methodology. Some issues, however, cannot be reviewed at all, and some arguments rarely succeed. Successful challenges are built on a clear understanding of the rules, a careful review of the contractor's documentation, qualified statistical analysis, and strict compliance with appeal procedures and deadlines.

This article explains the legal framework for challenging extrapolation, the two main lines of attack, the arguments that tend to work and those that generally do not, what information to request, how to work with a statistical expert, how appeals are handled at each level, and how to protect against recoupment while you appeal. It is educational and not legal advice. Providers facing extrapolated overpayments should work with experienced healthcare counsel and a qualified statistician.

The Legal Framework

When Extrapolation Is Allowed

Under Section 1893(f)(3) of the Social Security Act, a Medicare contractor may not use extrapolation to determine overpayment amounts unless the Secretary determines that there is a sustained or high level of payment error or that documented educational intervention has failed to correct the payment error.

What Cannot Be Reviewed

The same provision states that there shall be no administrative or judicial review of determinations of sustained or high levels of payment error. The Medicare Appeals Council has held that an ALJ erred as a matter of law by invalidating a statistical sample on the ground that the contractor failed to establish a sustained or high level of payment error, and federal courts have similarly held that they lack jurisdiction to review that determination. As a practical matter, providers should generally focus their challenges elsewhere.

The Presumption of Validity

The Medicare Appeals Council has explained that a statistical sample creates a presumption of validity and that the burden then shifts to the provider to attack the statistical validity of the sample or to challenge the determinations on the individual sampled claims. CMS Ruling 86-1, which sets out CMS policy on the use of sampling and extrapolation, is binding on ALJs and the Council.

What the Manual Says About Contractor Errors

CMS's Medicare Program Integrity Manual, Chapter 8, provides detailed instructions for statistical sampling. However, the manual states that failure by a contractor to follow its sampling instructions may prompt CMS to review the contractor's performance but does not, by itself, invalidate the sample or the projection. To succeed, a challenge generally must show that the sample is actually statistically invalid or that there was an actual error that affected the overpayment amount.

Two Ways to Reduce an Extrapolated Overpayment

1. Challenge the Individual Sample Claims

The most reliable way to reduce an extrapolated overpayment is to overturn denials on sampled claims. Because the extrapolation projects the sample results to the universe, each overturned claim reduces the average overpayment and, in turn, the projected amount. If enough sample claims are overturned, the error rate may fall significantly, and the extrapolated amount may decline sharply.

Claim-level challenges focus on whether each sampled service met Medicare requirements, such as:

• Whether the medical record supports medical necessity under applicable coverage policies.

• Whether required orders, certifications, plans of care, and signatures were present.

• Whether the contractor applied the correct coverage rules in effect on the date of service.

• Whether the reviewer misread or overlooked documentation.

• Whether the provider should be held liable under the limitation on liability or found without fault.

CMS's manual notes that when a liability waiver applies or the provider is found without fault for a sampling unit, the overpayment for that unit is treated as zero in the estimate.

2. Challenge the Statistical Methodology

The second approach is to challenge how the sample was designed, selected, or projected. Methodology challenges require careful review by a qualified statistician and are most effective when they identify concrete, material errors.

Methodology Arguments That May Have Merit

The following issues may support a challenge, depending on the facts. Each should be evaluated by a qualified statistical expert.

The Sample Cannot Be Replicated

CMS's manual requires contractors to document the random number source, software, starting point, and seed values so that the sample can be replicated, and to maintain documentation sufficient to allow replication and validation by an administrative or judicial body. If the contractor's documentation does not allow the sample to be recreated, that may support an argument that the methodology cannot be validated.

Errors in the Universe or Sampling Frame

The sampling frame should accurately represent the universe of claims under review. Potential problems include:

• Including claims that were not paid or that were outside the review period.

• Including services that do not meet the review criteria.

• Excluding claims that should have been included.

• Duplicate entries.

• Mismatches between the frame and the universe described in the demand letter.

The manual explains, for example, that unpaid sampling units are excluded from the frame for certain claim types. Errors that change which claims could be selected, or how the projection is calculated, may affect the validity of the estimate.

Inconsistent or Inappropriate Sampling Units

If the sampling unit is defined inconsistently, such as mixing claim lines and full claims, or if the projection method does not match the sampling unit, the estimate may be flawed.

Problems With Stratification

When stratified sampling is used, errors in defining strata, assigning claims to strata, or weighting the results may affect the estimate. A statistician can evaluate whether the stratification and estimation formulas were applied correctly.

Calculation Errors

Mathematical errors in computing the point estimate, standard error, or lower limit of the confidence interval can directly affect the demand amount. CMS's manual generally requires the demand to be based on the lower limit of a one-sided 90 percent confidence interval and allows the point estimate only where there are sound reasons and high precision. A statistician should recompute the estimate from the contractor's data.

Failure to Account for Underpayments

CMS's manual states that underpayments in the sample are treated as negative overpayments. If the contractor ignored underpayments, the estimate may be overstated.

Failure to Recalculate After Appeals

When sample claims are overturned at any appeal level, the extrapolated amount should be recalculated using the revised sample results. CMS's manual addresses changes in the overpayment estimate resulting from appeals. Providers should confirm that recalculation occurs after each favorable decision.

Missing Statistical Expert Review

CMS's manual requires a statistician or person with equivalent expertise to review and approve the methodology in writing, and to review results before the findings or demand letter are issued. It also sets minimum qualifications for statistical experts. If the required review did not occur, that fact may be relevant, although it may not by itself invalidate the sample unless it led to an actual error.

Use of the Point Estimate Without Justification

If the contractor demanded the point estimate instead of the lower limit, the provider can ask whether the manual's conditions for doing so were met.

Arguments That Generally Do Not Succeed

"The Sample Is Too Small"

CMS's manual states that it is neither possible nor desirable to specify a minimum sample size and that a small sample does not by itself invalidate an estimate if probability sampling and estimation were properly performed. The Medicare Appeals Council has called a challenge based solely on sample size without merit. Small sample size may be relevant to precision, but it is generally not enough on its own.

"There Was No Sustained or High Level of Payment Error"

As discussed above, the statute bars administrative and judicial review of this determination.

"The Contractor Did Not Follow the Manual"

Without showing an actual statistical problem or an error that affected the overpayment amount, manual noncompliance generally does not invalidate the sample.

"Extrapolation Is Unfair"

General objections to extrapolation as a method are unlikely to succeed, because statistical sampling is an established method recognized in CMS policy and by courts.

What to Request From the Contractor

To evaluate a methodology challenge, providers and their statistical experts need complete information. CMS's manual requires demand letters and final results to explain the sampling methodology, including the universe, frame, sampling unit, selection procedure, strata and sample sizes, period, estimation method, sampling error, and overpayment amounts for each sampled claim, and requires contractors to maintain detailed documentation.

Providers should review the demand letter and request, as appropriate:

• The electronic sampling frame and universe.

• The sampling plan and methodology description.

• The random number generation details, including software, seed, and starting point.

• Stratification details, if used.

• The list of sampled claims with paid amounts and overpayment determinations.

• Reviewer worksheets and claim-level rationales.

• Overpayment and underpayment worksheets.

• The statistical expert's approval and calculations.

Make requests in writing and keep copies. If information is missing, document that fact.

Working With a Statistical Expert

A qualified statistician is essential for any methodology challenge. The expert can:

• Review whether the sample was a valid probability sample.

• Attempt to replicate the sample selection.

• Compare the sampling frame with the universe.

• Recompute the estimate and confidence interval.

• Evaluate stratification and estimation formulas.

• Identify the effect of any errors on the overpayment amount.

• Prepare a written report or declaration for the appeal.

• Testify at an ALJ hearing if needed.

Choose an expert with experience in Medicare statistical sampling. The expert's report should explain clearly what was wrong, why it matters statistically, and how it affects the overpayment amount. Generic critiques are much less persuasive than specific, quantified findings.

The Appeals Process and Key Deadlines

Extrapolated overpayments are appealed through the Medicare fee-for-service appeals process:

1. Redetermination by the MAC, generally within 120 days of receiving the demand letter or initial determination.

2. Reconsideration by a Qualified Independent Contractor (QIC), generally within 180 days of the redetermination decision.

3. ALJ hearing at the Office of Medicare Hearings and Appeals, generally within 60 days of the reconsideration decision, if the amount in controversy is met.

4. Medicare Appeals Council review, generally within 60 days of the ALJ decision.

5. Judicial review in federal district court, if the amount in controversy is met.

Raise Methodology Challenges Early

Providers should raise both claim-level and methodology challenges early, starting at redetermination and reconsideration, and submit supporting evidence, including the statistical expert's report, by the reconsideration level. Under Medicare appeals rules, providers generally must show good cause to submit new evidence at the ALJ level that was not submitted before the QIC reconsideration.

Special Requirements at the ALJ Level

Federal regulations contain specific requirements for ALJ appeals involving statistical sampling:

• Under 42 CFR 405.1014(a)(3), if the appellant disagrees with how a statistical sample or extrapolation was conducted, the appellant must include the required information for each sample claim it wishes to appeal, file the request for hearing for all sampled claims it wishes to appeal within 60 calendar days of receiving the last reconsideration if the claims were not all addressed in one reconsideration, and assert the reasons it disagrees with how the statistical sample or extrapolation was conducted in the request for hearing.

• Under 42 CFR 405.1032(d), if the request for hearing does not assert reasons for disagreeing with how the statistical sample or extrapolation was conducted, issues related to the sampling and extrapolation will not be considered or decided. When those issues are raised, the ALJ or attorney adjudicator must base the decision on a review of the entire sample to the extent appropriate to decide the issue.

• Under 42 CFR 405.1064, when an appeal from the QIC involves an overpayment issue and the QIC used a statistical sample, the ALJ must base the decision on a review of the entire statistical sample used by the QIC.

These rules make it essential to state methodology objections clearly in the hearing request and to coordinate appeals of all sampled claims.

OMHA Statistical Sampling Initiative

Separately, the Office of Medicare Hearings and Appeals has offered a Statistical Sampling Initiative that allows certain appellants with large numbers of pending ALJ appeals to have a sample of those appeals adjudicated and the results extrapolated to the full set. This is a different use of sampling, initiated by the appellant, and is not a challenge to a contractor's extrapolation. Providers considering it should review OMHA's current eligibility rules carefully.

Protecting Against Recoupment While You Appeal

Under Section 1893(f)(2) of the Social Security Act and 42 CFR 405.379, Medicare generally may not recoup an overpayment while a timely redetermination or reconsideration is pending. To obtain this protection, providers generally must:

• File the redetermination within 30 days of the date of the demand letter.

• File the reconsideration within 60 days of the redetermination decision.

Filing within the standard 120-day or 180-day windows preserves appeal rights but may not prevent recoupment. After the reconsideration level, recoupment may begin even if an ALJ hearing is requested, and interest continues to accrue. If the provider ultimately prevails, interest may be payable to the provider on amounts recouped under applicable rules.

Providers that cannot repay immediately may request an extended repayment schedule from the MAC.

A Step-by-Step Plan for Challenging an Extrapolated Overpayment

1. Calendar every deadline from the demand letter, including the 30-day redetermination deadline to limit recoupment.

2. Engage experienced healthcare counsel and a qualified statistician immediately.

3. Obtain the complete sampling documentation and claim-level findings.

4. Review every sampled claim for appealable issues, organizing medical records and supporting policy references.

5. Have the statistician evaluate the methodology, attempt replication, and recompute the estimate.

6. Decide which methodology arguments are supported by evidence and material to the overpayment amount.

7. File a timely redetermination with claim-level arguments, methodology objections, and supporting evidence.

8. File a timely reconsideration, adding any new evidence and the statistician's report.

9. If needed, request an ALJ hearing that addresses all sampled claims you wish to appeal and clearly states the reasons you disagree with the sampling and extrapolation.

10. Confirm that the overpayment is recalculated after each favorable decision.

11. Address any related 60-day overpayment obligations for periods outside the sample, and correct the underlying compliance problems.

A Hypothetical Example

A home health agency receives an extrapolated overpayment demand of $780,000 based on a sample of 60 claims, 30 of which were denied. The agency immediately calendars the 30-day redetermination deadline and engages counsel and a statistician.

The claim-level review shows that 12 of the 30 denials involved homebound status documentation that was actually present in the records but was overlooked by the reviewer, and that four others were denied under a coverage policy that was not in effect on the dates of service. The statistician's review finds that the sampling frame included 85 claims that were never paid and that the contractor did not count two underpaid claims as negative overpayments.

The agency files a timely redetermination and reconsideration with organized records for each sample claim, policy references, and the statistician's report quantifying the effect of the frame and underpayment errors. At the reconsideration level, 14 of the denials are reversed and the frame is corrected. The recalculated overpayment drops substantially. The agency continues to appeal the remaining claims it believes are supported, while also fixing its homebound documentation practices and evaluating whether it has related overpayment obligations for periods outside the sample.

This example illustrates how combining specific claim-level arguments with concrete, quantified methodology findings can produce a much better result than general objections to extrapolation.

Practical Tips

• Prioritize claim-level appeals. They are often the most effective way to reduce the extrapolated amount.

• Be specific. Quantify the impact of each methodology error.

• Do not rely on sample-size arguments alone.

• Do not wait until the ALJ level to raise methodology issues or submit evidence.

• Keep records organized by sample claim number.

• Track the recalculated overpayment after each decision.

• Consider settlement or alternative resolution options where appropriate, with counsel's advice.

• Use the experience to improve documentation and reduce future error rates.

Frequently Asked Questions

Can I challenge the contractor's finding of a high error rate?

Generally no. Section 1893(f)(3) of the Social Security Act bars administrative and judicial review of determinations of sustained or high levels of payment error.

What is the most effective way to reduce an extrapolated overpayment?

Overturning individual sample claims is often the most effective approach, because each overturned claim reduces the projected overpayment.

Do I need a statistician?

For a methodology challenge, yes. A qualified statistical expert can evaluate the sample design, replicate the selection, and recompute the estimate.

When should I raise methodology issues?

As early as possible, beginning at redetermination and reconsideration. At the ALJ level, you must state your reasons for disagreeing with the sampling and extrapolation in the request for hearing, or those issues will not be considered.

Will the overpayment be recalculated if I win some claims?

It should be. The extrapolated amount is generally recalculated based on the revised sample results after favorable appeal decisions.

Conclusion

Challenging a Medicare statistical sample and extrapolation requires strategy and precision. Providers cannot challenge the sustained or high level of payment error determination, and arguments based only on sample size or manual noncompliance rarely succeed. Effective challenges focus on overturning individual sample claims and on demonstrating specific, material statistical errors, such as an unreplicable sample, a flawed sampling frame, calculation mistakes, or failure to recalculate after appeals. By acting quickly, obtaining complete documentation, working with experienced counsel and a qualified statistician, raising issues early, meeting every deadline, and protecting against recoupment, providers can significantly improve their chances of reducing an extrapolated overpayment.

How HealthBridge Can Help

HealthBridge is a healthcare management consulting firm that helps providers respond to extrapolated Medicare overpayment demands. We organize sample claim documentation, analyze contractor findings, coordinate with statistical experts and legal counsel, prepare appeal packages, track deadlines and recalculations, and help implement corrective actions to reduce future risk. If you have received an extrapolated overpayment demand, please get in touch with us. We will be happy to help you with this topic.

References

• U.S. Social Security Administration. Social Security Act, Section 1893, Medicare Integrity Program. https://www.ssa.gov/OP_Home/ssact/title18/1893.htm

• Office of the Law Revision Counsel, U.S. House of Representatives. 42 U.S.C. 1395ddd, Medicare Integrity Program. https://uscode.house.gov/view.xhtml?req=granuleid:USC-prelim-title42-section1395ddd&num=0&edition=prelim

• Centers for Medicare & Medicaid Services. Medicare Program Integrity Manual, Chapter 8, Section 8.4, Use of Statistical Sampling for Overpayment Estimation. https://www.cms.gov/regulations-and-guidance/guidance/manuals/downloads/pim83c08.pdf

• Centers for Medicare & Medicaid Services. CMS Ruling 86-1. https://www.cms.gov/regulations-and-guidance/guidance/rulings/downloads/cmsr8601.pdf

• U.S. Department of Health and Human Services, Departmental Appeals Board. Medicare Appeals Council Decision, Lakeside Foot Clinic (October 15, 2009). https://www.hhs.gov/sites/default/files/static/dab/decisions/council-decisions/lakeside_footclinic.pdf

• U.S. Department of Health and Human Services, Office of Medicare Hearings and Appeals. Statistical Sampling Initiative Fact Sheet. https://www.hhs.gov/sites/default/files/ST-4256_OHMA%20website%20Stat%20sample%20Fact%20Sheet%206-2_remediated%20.pdf

• U.S. Department of Health and Human Services, Office of Inspector General. RAT-STATS Statistical Software. https://oig.hhs.gov/compliance/rat-stats/

• Electronic Code of Federal Regulations. 42 CFR 405.1014, Request for an ALJ Hearing or a Review of a QIC Dismissal. https://www.ecfr.gov/current/title-42/section-405.1014

• Electronic Code of Federal Regulations. 42 CFR 405.1018, Submitting Evidence. https://www.ecfr.gov/current/title-42/section-405.1018

• Electronic Code of Federal Regulations. 42 CFR 405.1032, Issues Before an ALJ or Attorney Adjudicator. https://www.ecfr.gov/current/title-42/section-405.1032

• Electronic Code of Federal Regulations. 42 CFR 405.1064, ALJ Decisions Involving Statistical Samples. https://www.ecfr.gov/current/title-42/section-405.1064

• Electronic Code of Federal Regulations. 42 CFR 405.379, Limitation on Recoupment of Provider and Supplier Overpayments. https://www.ecfr.gov/current/title-42/section-405.379

• Electronic Code of Federal Regulations. 42 CFR 401.305, Requirements for Reporting and Returning of Overpayments. https://www.ecfr.gov/current/title-42/chapter-IV/subchapter-A/part-401/subpart-D/section-401.305

• Centers for Medicare & Medicaid Services. Original Medicare (Fee-for-Service) Appeals. https://www.cms.gov/medicare/appeals-grievances/fee-for-service