On September 29, 2015, the United States Court of Appeals for the Fourth Circuit agreed to review whether statistical sampling can be used to prove liability under the False Claims Act without directly analyzing the underlying Medicare billing claims (United States ex rel. Michaels et al. v. Agape Senior Community Inc. et al.). This will be the first appellate ruling on this hotly contested tactic.
Relators and government agencies highly favor statistical sampling because the practice allows them to prove the falsity of only a small number of claims and then extrapolate the falsity findings to a larger universe of claims, thereby potentially exponentially increasing available damages and penalties. Allowance of statistical sampling typically results in greater pressure to settle and also may result in larger settlements.
A Tennessee federal district court was one of the first to allow the use of extrapolation to show False Claims Act liability (United States ex rel. Martin v. Life Care Centers of America, Inc.). During 2015, some other district courts followed suit, with the caveat that defense could attack the reliability of sampling later on.
The district court in Michaels, however, rejected the use of statistical sampling to prove liability. The relators sought interlocutory review of that opinion. The relators accused the defendants of overbilling Medicare for nursing home services. The district court held that statistical sampling would be permissible only if it were impossible, rather than time-consuming and expensive, to directly analyze billing claims.
In general, although statistical sampling may be useful in determining damages once liability has been established, it is not appropriate to prove liability itself. A relator’s burden of proof requires that each of the essential elements for each false claim be proven by a preponderance of the evidence. Thus, without the presentation of factual evidence relating to each claim for which a relator intends to establish liability, the relator cannot satisfy this burden of proof.
This is especially true in medical necessity cases because for each alleged false claim, the relator must show that the medical treatment in question was objectively unreasonable. For the claims to have been false, at minimum, the treatment had to be improper — or, in the terms of the statute, not reasonable and necessary. See 42 U.S.C. § 1395y(a)(1)(A). Determining the appropriate treatment a patient should receive is a complex judgment call made by physicians on the scene. Actions alleging Medicare/Medicaid fraud typically involve issues of individualized medical decision making. As a result, it is not possible to extrapolate those findings from a sample to a universe rife with variability.
Any decision out of the Fourth Circuit will go a long way toward providing clarity on this unsettled and highly disputed method for proving liability.