Whether and to what extent statistical sampling and extrapolation may be used to establish liability in FCA cases will soon be decided by the Fourth Circuit and the decision is sure to have far-reaching repercussions for how FCA cases are brought and defended. 

While the use of statistical sampling and extrapolation to determine damages once liability has already been established is common practice in FCA cases, the increasing reliance on these questionable methods as the exclusive means of proving liability – especially in medical judgment cases that depend on a careful case-by-case analysis of each patient’s particular situation – marks a troubling new trend in FCA litigation. Indeed, the consequences to health care providers of allowing relators to establish liability based on statistical extrapolation would be difficult to overstate. 

Case on Review

In the case on review, the relators alleged that the defendant’s claims for hospice services were “false or fraudulent” because the patients were ineligible for hospice treatment or at least for the level of hospice treatment they received. After the government declined to intervene, the whistleblowers pursued the litigation and sought to prove not only damages, but the falsity of the claims through statistical sampling, arguing that it could potentially cost well over $30 million to have their experts review each of the individual claims at issue. The district could rejected the use of statistical sampling, finding that each claim asserted presented the question of whether certain services furnished to nursing home patients were medically necessary, and that answering that question for each of the patients involved was a highly fact-intensive inquiry requiring medical testimony after a thorough review of the detailed medical charts. While some cases are uniquely suited for statistical sampling, the court noted, this was not such a case. The court nevertheless certified its ruling for immediate review in light of the disagreement among the district courts with respect to the appropriateness of statistical sampling. 

Potential Impact

A reversal of the district court’s decision could have a crippling impact on FCA defendants, particularly in health care cases, which typically involve a large universe of potentially false claims, and even more particularly in medical judgment cases, where statistical methodology is used as a substitute for claim-by-claim and patient-by-patient evidence. Until the issue is decided by the circuit courts, FCA defendants must continue to operate under the assumption that the government and relators may satisfy their burden of proof on liability using statistical sampling and extrapolation and must be prepared to protest the need for statistical methodologies by showing, among other things, that the medical charts are intact and available for review by both parties and that there exist discrete claims that may be analyzed and subjected to cross-examination. In cases where statistical sampling may be necessary, FCA defendants must be prepared to mount Daubert challenges to the reliability of the government’s methodology and to develop their own sampling and extrapolation methodologies. 

The Court will also consider whether the government has an absolute, unreviewable veto over the settlement of FCA cases in which the government has declined to intervene.