Courts have generally accepted statistical sampling and extrapolation as evidence to estimate damages.[i] However, the use of sampling for purposes of establishing liability in False Claims Act (“FCA”) litigation is an issue of conflict between those prosecuting FCA claims and those defending such claims. In the decision in United States ex rel. Martin v. Life Care Centers of America, Inc., the Eastern District of Tennessee found sampling and extrapolation appropriate for the Government’s use in attempting to establish the elements of FCA liability.[ii] While that case settled for $145 million in 2016, this post offers an overview of the case, along with a summary of the Court’s decision and a discussion of its implications…

Overview of Life Care and FCA Sampling

The Government alleged that Life Care, a provider of skilled nursing services comprised of more than 200 facilities, submitted fraudulent claims to Medicare as part of a nationwide scheme to provide medically unnecessary services. The Government planned to present specific evidence as to a sample of 400 patient admissions. With that sample, it planned to extrapolate both liability and damages as to the additional universe of 54,396 unidentified patient admissions, for which it did not plan to provide any specific evidence.  Life Care sought summary judgment as to the unidentified claims, arguing that the Government could not satisfy its burden of proving knowing falsity as to the unidentified claims merely by extrapolation and that allowing it to do so would violate Life Care’s right to due process.[iii] The Government argued in response that the use of statistical evidence is generally accepted in other types of litigation and has been used previously in the FCA context.[iv]  On September 29, 2014, the Court denied Life Care’s motion, finding, in part, that Life Care’s objections were insufficient to bar the Government’s use of statistical sampling, and that arguments regarding the reliability and limitations of statistical sampling would be better considered by a jury.[v]

Life Care Court Upholds Sampling for Liability

The Court found no precedent determinative of whether extrapolation could be used to prove FCA liability in a case involving Medicare overpayments.[vi]  In their briefs, both parties highlighted cases in which they argued courts had either permitted or declined to use statistical sampling to prove claims brought under the FCA.  Additionally, the parties presented the Court with a variety of other rulings from civil, criminal and administrative actions addressing the use of statistical sampling.  In its analysis, the Court found each of these cases to be distinguishable from Life Care.  For example, the Government’s argument relied heavily on FCA cases and criminal cases in which statistical sampling and extrapolation was permitted to establish loss or damages.[vii]  However, the Court noted that “using extrapolation to establish damages when liability has been proven is different than using extrapolation to establish liability.”[viii]

Without finding determinative precedent, the Court turned to analyzing whether extrapolation evidence is generally applicable to each element of the FCA claims against Life Care—evidence of specific claims, falsity, knowledge, and materiality. The Court also addressed the defendant’s arguments regarding due process

Considerations for Life Care

Although the Court did not preclude the Government from using extrapolation evidence to support its case, it did leave open the opportunity for Life Care to argue to the jury that the Government’s process and conclusions are flawed. Thus, Life Care may be best suited to focus its arguments on any potential weakness and technical imprecision of the extrapolation analysis.

When evaluating a sampling and extrapolation analysis, particularly one involving potentially fraudulent claims, a variety of missteps could result in an invalid analysis or lead to overly imprecise conclusions.

Implications of the Life Care Ruling

The District Court’s ruling is unlikely to be the final decision on the topic of sampling and extrapolation in FCA matters. However, it could provide a foothold for the DOJ to promote further acceptance of the use of sampling and extrapolation to establish liability in FCA cases.  If the Eastern District of Tennessee’s view is adopted by other courts, the cost and strategy of prosecuting and defending FCA litigations may be dramatically impacted.  For the Government, the ability to employ sampling and extrapolation could reduce the need to investigate and scrutinize large volumes of individual claims, thereby reducing the cost and effort required to bring FCA litigation, while potentially increasing the estimated damages and penalties at stake. Conversely, the cost to defendants could swell significantly if they must investigate and prepare defenses to expanded universes of claims that have not been identified by the Government.  In addition, if courts approve the practice of extrapolating liability for investigated claims company-wide to include uninvestigated, unidentified claims, it could dramatically increase the number of claims at issue in the average FCA case.

Although the Court denied Life Care’s motion in this matter, it may have also indicated the path toward successful motions in the future. As the Court noted in its ruling, “as long as the statistical sample is a valid sample that is representative of the universe of claims, the natural disparity between the claims does not preclude using sampling and extrapolation as evidence of the total number of claims for non-covered services.”[ix]  This language suggests that defendants who can successfully argue that a sample is not representative, or otherwise invalid, may succeed on motions for summary judgment in similar circumstances.

With these types of arguments in play, FCA defendants will likely benefit from engaging statisticians and forensic experts much earlier in the litigation process to rebut Government claims and prepare their own affirmative defenses. When developing those defenses, arguments involving statistical theory alone may be less successful given the need for applied statistics and analysis that is designed and implemented with the broader regulatory environment in mind. Perhaps more importantly, this ruling increases the likelihood that juries will hear and decide the merit of complex statistical evidence, so experts and statisticians with testimony and communication expertise may play an increasingly prominent role in the matter.


The ruling on Life Care’s motion for summary judgment could set the stage for a dramatic change in the landscape of litigating and defending False Claims Act cases. Deciding that the Government may use sampling and extrapolation in an FCA case to attempt to establish liability for unidentified claims may directly impact the way companies, attorneys and their experts develop strategies to litigate and defend claims under the FCA.


[i] See, e.g., United States v. Maceo, 873 F.2d 1 (1st Cir. 1989) (upholding the use of sampling in criminal trial); Blue Cross & Blue Shield of N.J. v. Philip Morris, Inc., 113 F. Supp. 2d 345, 372-75 (E.D.N.Y. 2000) (accepting sampling in civil RICO action).
[ii] United States ex rel. Martin v. Life Care Ctrs. of Am., No. 08-cv-251, Dkt. 184 (E.D. Tenn. Sept. 29, 2014) (“Opinion”).
[iii] Id. at Dkt. 141 (“Life Care’s Motion”).
[iv] Id. at Dkt. 152 (“Government’s Brief”).
[v] Opinion at 38.
[vi] Id. at 25.
[vii] See United States v. Jones, 641 F.3d 706, 712 (6th Cir. 2011); United States v. Rogan, 517 F.3d 449, 453 (7th Cir. 2008); United States v. Fadul, 2013 WL 781614 at *14 (D. Md. 2013).
[viii] Opinion at 22.
[ix] Opinion at 29 (emphasis added).

This post was written by Christopher Haney, CPA, CFE, CHC, and portions were originally published in Law360’s Expert Analysis in October 2014.

This post does not, in any way, constitute legal advice, nor does it offer guidance or legal opinions about how courts or other bodies may interpret particular issues of statistical or financial analysis. The cases cited are merely examples of relevant issues, and they are included to demonstrate how the issues were treated based on the facts and circumstances of each specific case.