There is little dispute that the False Claims Act (FCA) is among the most potent weapons for fighting fraud and government waste. In just this past fiscal year alone, the Department of Justice (DOJ) utilized the FCA to recover more than $3.7 billion across a diverse array of industries such as health care, housing, and government procurement. Recoveries in individual actions reached well into the hundreds of millions of dollars, including a whopping $465 million settlement on behalf of the federal government and state Medicaid programs as a result of one health care company's misclassification of a generic drug.

Naturally, such recoveries under the FCA involve a tremendous number of false claims. In these kinds of large-scale matters, the government and qui tam relators alike have relied upon new and improved methods to track and prosecute their cases. Principal among these strategies is statistical sampling and extrapolation: wherein a plaintiff identifies a representative sample of claims and projects inferences from an analysis of those claims on to all of the claims at bar. Defendants often object to the use of statistics in this fashion on the grounds that it is at odds with the FCA and their right to raise individualized defenses to individual claims.

Unfortunately, trial courts do not have the benefit of clear or controlling authority for resolving these disputes. Neither the statutory text nor an opinion from a circuit-level court provides express guidance on the viability of statistical sampling under the FCA. In the absence of such guidance, courts have come down on this issue in a variety of ways, with each continuing to make case-by-case determinations premised on their “responsibility to determine the fairest course of action based upon the facts presented and the claims asserted.” See United States. v. Agape Senior Community, No. 12-3466, (D.S.C. June 25, 2015) (Agape I), order corrected, No. 12-3466, (D.S.C. July 6, 2015), and aff'd in part, appeal dismissed in part sub nom. United States v. Agape Senior Community, 848 F.3d 330 (4th Cir. 2017). This article combines perspectives from relator and defense counsel to provide measured insight into some of those decisions, the current legal landscape, and where the law may be heading.

The State of Statistics in FCA Cases

When a court must determine the viability of statistical sampling in a FCA case, the first question it must ask is what the plaintiffs are proffering such evidence to prove: damages or liability?

Generally speaking, courts have been more willing to allow plaintiffs to use statistical sampling and extrapolation techniques to establish damages where liability has already been proven. See, e.g., United States v. Vista Hospice Care, No. 07-00604, at *11 (N.D. Tex. June 20, 2016); United States v. Life Care Centers of America, 114 F. Supp. 3d 549, 560 (E.D. Tenn. Sept. 29, 2014). But, this issue is not completely settled, as at least one court has deemed the use of statistical sampling for damages purposes inappropriate, see United States v. Friedman, No. 86-610, 1993 U.S. Dist. LEXIS 21496, at *9 n.1 (D. Mass. July 23, 1993).

FCA plaintiffs often face a harder row on the liability front, with many more courts refusing to permit statistical sampling due to, inter alia, the individualized nature of the claims in suit. See, e.g., Vista Hospice Care; United States v. Medco Physicians Unlimited, No. 98-1622, 2000 U.S. Dist. LEXIS 5843, at *23 (N.D. Ill. Mar. 15, 2000). Yet, as with damages, courts are not unanimous here either; indeed, numerous decisions have authorized the use of statistical sampling in order to establish falsity under the FCA. See United States v. Rogan, 517 F.3d 449, 452-53 (7th Cir. 2008); United States v. Robinson, No. 13-27 (E.D. Ky. Mar. 31, 2015).

While the various reasons for and against statistical sampling can be gleaned from a review of current precedent, the Life Care opinion contains a particularly thorough treatment of the debate. Beginning with the recognition that “statistical sampling has been generally limited to determine damages, rather than liability,” Life Care, 114 F. Supp. 3d at 560, the court went on to explain that “statistical sampling to find liability for extrapolated claims could be in conflict with the government's burden to establish the elements of a FCA claim.” Despite this possibility, the Life Care court ultimately admitted the statistical evidence at issue because proving the falsity of each individualized claim “would [have] consume[d] an unacceptable portion of the court's limited resources.” In so holding, however, the court emphasized that the evidence's admission did not define the weight it was to be afforded; aptly noting that its “ruling today simply holds that statistical sampling may be used to prove claims brought under the FCA involving Medicare overpayment, but it does not and cannot control the weight that the fact finder may accord to the extrapolated evidence … rather, the burden of determining the weight of the evidence lies with the fact finder.”

Where to Next?

Last year, the FCA bar waited for the Fourth Circuit's ruling in United States v. Agape Senior Community, 848 F.3d 330 (4th Cir. 2017) (Agape II), with the hope that they would finally receive some guidance on the suitability of statistical sampling and extrapolation. Unfortunately, the Fourth Circuit punted on the statistical evidence issue; refusing to decide whether statistical sampling and extrapolation may, as a matter of law, be used to demonstrate liability or damages in a FCA case. The court ultimately held that the issue, as presented by the appellants, did not constitute a “pure question of law” appropriate for interlocutory review under Section 1292(b). On that basis, the Fourth Circuit ruled that it was “constrained to dismiss that aspect of the relators' appeal as improvidently granted.”

Other courts have, however, recently addressed the use of statistical sampling and extrapolation in different areas of the law. For example, in the year before Agape II, the Supreme Court decided Tyson Foods v. Bouaphakeo, 136 S. Ct. 1036 (2016), in which it examined the use of statistics in determining liability and damages under the Fair Labor Standards Act. Similar to the Life Care court, Justice Anthony Kennedy ultimately declined to categorically exclude the use of statistical evidence and affirmed its admission by the lower court in light of the circumstances of that case. This holding may very well provide a roadmap for courts faced with similar issues within the FCA context; but, for now, we are left to wait and see.

Conclusion

Whether, how, and to what extent statistical sampling and extrapolation can be used to establish FCA liability and damages are issues that remain in a precedential abyss. In light of the ever-increasing size and scope of FCA actions, the issue will only continue to grow in importance. Only time—and clear appellate guidance—will tell where this ongoing debate goes.

Benjamin H. McCoy is an attorney with Fox Rothschild. He has a broad commercial practice with an emphasis on international business and healthcare litigation. He can be reached at [email protected].

Zac Arbitman is an attorney at Kessler Topaz Meltzer & Check, where he prosecutes complex antitrust cases, consumer class actions, and whistleblower matters. He can be reached at [email protected].