FCA statistical sampling and extrapolation expert

Expert class library · Healthcare fraud and False Claims Act · Either side

Designs or critiques samples of claims, loans or patient files and extrapolates findings to the whole universe to estimate overpayments and False Claims Act damages. Courts usually admit a properly drawn random sample and treat sample size disputes as weight, but they exclude damages extrapolations built on unsupported assumptions or on samples with no statistical basis.

What this expert is retained to answer

• How was the sample drawn, was it random and stratified, and who agreed to the design?
• Is the sample large enough to support extrapolation to the whole universe of claims, and did anyone test its statistical validity?
• What assumption turns a statistical pattern into a dollar figure, and is that assumption grounded in the facts of individual claims?
• Are the reviewers who scored the sampled claims applying the governing payment or underwriting standards?
• Does the statistician claim to prove fraud, or only a pattern that other evidence must tie to false claims?

Methods

• Probability sample design (RAT-STATS)
• Stratified random sampling
• Confidence interval and lower bound analysis
• Sample size adequacy
• Regression of billing patterns

How the testimony is attacked

• Waived design challenges. Defendants who negotiated the sampling protocol may lose later objections to it. In United States v. Allied Home Mortgage the Fifth Circuit held that challenges to the sample other than its size were waived because the defendants had agreed to the methodology in discovery.
• Sample size. Defendants argue the sample is too small to support extrapolation. Allied treated the preserved sample size objection as going to weight, and in Grant v. Zorn the Eighth Circuit upheld admission in a bench trial of a CPA's extrapolation from thirty-one patient files although no one calculated the sample's statistical validity.
• Damages built on an unsupported assumption. A reliable statistical pattern does not make every damages model reliable. In United States v. Long Grove Manor the court admitted a statistician's regression showing therapy spikes on assessment days but excluded his overpayment estimates, which assumed without basis that the lower therapy level reflected each patient's true needs.
• Non-random probe samples. Opinions that generalize from a handful of hand-picked files invite exclusion. In Long Grove Manor the court barred a physician from inferring a facility-wide fraud scheme from a non-random review of four patient charts.
• Statistics as proof of fraud. Defendants argue that pattern evidence cannot show any particular claim was false. Long Grove Manor admitted the pattern analysis because the statistician did not claim to pass on the validity of specific claims, leaving liability to other proof.

What the public record shows

A deliberately narrow CourtListener search, (extrapolat* OR "statistical sampling") AND (Medicare OR "False Claims") AND expert AND (Daubert OR "Rule 702"), returned 13 opinions filed since 2015, as of October 3, 2026; broader searches return more. Three that show how courts handle this class of testimony:

Each case links to the free opinion text on CourtListener.

Under amended Rule 702

Since December 1, 2023, Rule 702 says expressly that the party offering an expert must show the court it is more likely than not that the testimony meets each requirement: that it rests on sufficient facts or data, uses reliable methods, and reflects a reliable application of those methods to the case. Questions about the basis of an opinion are no longer automatically matters of weight for the jury. For how the circuits have applied the amendment, see the Rule 702 tracker, which follows each court of appeals; for what that means for preparing or attacking this class of expert, see the guide on amended Rule 702.

Related classes

• Forensic accountant
• Insurance actuary
• Disparate-impact labor statistician

Guides

• Amended Rule 702 After Nearly Three Years: What Courts Exclude

Simulating this class on a matter

Supreme Mind simulates a class of expert, never a named individual, on your fact pattern and returns the likely opinion, the ranked cross-examination weaknesses, the Rule 702 attack surface and what it means for settlement. Read how it works, or book a demo.

Last reviewed October 3, 2026. How this page is built: rulings are found by searching court opinions on CourtListener, and each one is read in the opinion before it is summarised here; outcomes are labelled by what the court did with the expert's testimony. No individual expert is named. This page summarises public decisions for orientation and is not legal advice; read the opinion before relying on any ruling.