Debate About Statistical Sampling, DOJ Authority to Reject Settlements, Continues In Fourth Circuit

As we reported here, the Fourth Circuit is currently facing a unique case presenting the question of whether the government has an unfettered veto authority over FCA settlements and, if not, whether the district court erred in rejecting the government’s objections to a settlement.  See United States ex rel. Michaels v. Agape Senior Cmty., Inc., No. 15-2145 (4th Cir.).  Notably, the government’s objections were premised on the use of statistical sampling to establish FCA liability, adding to the dispute a critical issue that has been generating significant debate.

The relators and defendant hospice provider (“Agape”) previously filed briefs (discussed here) urging the Fourth Circuit to uphold their settlement notwithstanding DOJ’s objection.  In its response, DOJ encourages the Fourth Circuit to side with the Fifth and Sixth Circuits in adopting a plain meaning reading of the FCA’s settlement provision, 31 U.S.C.§ 3730(b)(1).  The language of this provision, the government points out, “contrasts markedly with more qualified language in other provisions of the FCA,” such as the subsection allowing the United States to intervene after previously declining upon a showing of “good cause.”  The government further argues that while the FCA allows the United States to settle a qui tam suit over the objections of the relator—so long as the court determines the settlement is “fair, adequate, and reasonable”—there is no analogous judicial review provision in the subsection governing dismissal by the relator.  As a result, the government concludes that “[h]ad Congress intended to require the United States to provide reasons for refusing to consent to a settlement—which a court could review for reasonableness—it would have said so . . . .”

In addition to statutory arguments, the government also makes a policy argument, that restricting DOJ’s authority to veto a proposed settlement will remove an important check on relators’ personal interests in profiting from the sale of an overly broad release.  The government rejects Agape’s contention that the government’s interests could be protected through mechanisms short of an unlimited veto authority, maintaining that such protections can only be effectuated by amending the text of the FCA.

DOJ also emphatically defends the role of statistical sampling to establish FCA liability.  The government argues that the district court’s decision that the relator could not use statistical sampling to establish FCA liability at trial reasonably has no bearing on whether the government can use statistical sampling for the distinct purpose of evaluating the settlement value.  The government also attacks the district court’s view that statistical sampling should not be used unless “the evidence has dissipated,” particularly where each claim at issue requires unique fact-intensive scrutiny.  DOJ maintains that the very purpose of statistical sampling is to draw inferences about a universe of differing claims, and it urges the Fourth Circuit to adopt the reasoning set forth in a lengthy 2014 District of Tennessee opinion upholding the use of statistical sampling to establish liability (as reported here).  According to the government, Federal Rule of Evidence 702 and established Daubert standards should act as safeguards against inappropriate sampling, rather than categorical rules prohibiting the use of sampling in certain contexts.

Agape filed a second brief addressing the use of statistical sampling.  Agape urges the Fourth Circuit not to permit the relators to “sidestep their burden of proof (and along with it, Agape’s due process rights) by using statistical evidence to establish liability.”  Because the FCA requires proof of a knowing submission of a false claim, Agape argues that statistical sampling allows relators to take a shortcut past the most fundamental elements of their burden of proof.  Agape emphasizes the complex clinical judgments implicated by each claim and notes that liability cannot be supported merely by identifying an erroneous determination of hospice eligibility; rather, relators must prove that a treating physician did not or could not have believed a patient was eligible.  Aggregate data, Agape insists, cannot identify what led to the physician’s diagnosis, and therefore it fails to adduce the requisite evidence of both falsity and scienter.  Agape distinguishes its case from other Fourth Circuit cases in which extrapolation was permitted to calculate losses, noting that these cases did not involve attempts to prove knowing misconduct.

A copy of the government’s brief and Agape’s brief can be found here and here.