Statistical Evidence and the False Claims Act


The False Claims Act[1] is a fraud statute; therefore, False Claims Act complaints must be pled with particularity,[2] identifying “the who, what, when, where, and how of an actual false claim” submitted to the government.[3] Whistleblowers without evidence of specific claims have tried to circumvent the rule with statistics, showing a likelihood that false claims were submitted to the government. The courts have repeatedly rejected these attempts, saying, e.g.:

Rule 9(b) does not permit a False Claims Act plaintiff merely to describe a private scheme in detail but then to allege simply … that claims requesting illegal payments must have been submitted, were likely submitted or should have been submitted.”[4]

In 2013, it appeared the First Circuit Court of Appeals might have cracked opened the door to the use of statistical evidence in False Claims Act cases. In In re Neurontin Marketing and Sales Practices Litigation,[5] the court allowed a statistical expert to testify on the issue of causation, to offer his opinion that Pfizer’s marketing of Neurontin for off-label uses led to increased improper use of the drug.[6]

In 2017, the First Circuit Court of Appeals slammed the door on a whistleblower who tried to sneak through the crack left by the Neurontin cases.

In United States ex rel. Booker et al. v. Pfizer, Inc.,[7] the whistleblowers had no evidence of actual false claims submitted to the government, but they argued that they could prove the case statistically, citing the Neurontin cases for support. The court rejected the comparison. It said the Neurontin cases did not stand for the proposition that the existence of false claims could be proven through statistics. Rather:

In those cases, we held that plaintiffs could use aggregate data together with strong circumstantial evidence to overcome summary judgment on the distinct issue of whether there was a causal link between fraudulent marketing and demonstrated off-label prescriptions in the distinct context of a civil RICO case — not that such proof could be used to demonstrate the existence of false claims in an FCA case.[8]

The court’s clarification and limitation of the statistical discussion in the Neurontin cases should come as a welcome relief to government contractors, including health care providers, who face under-informed bounty hunters[9] hoping to strike it rich with the False Claims Act.

If you have any questions, please contact David B. Honig at or (317) 977-1447 or your regular Hall Render attorney.

[1] 31 U.S.C. § 3729-3733.

[2] Fed.R.Civ.P. 9(b).

[3] United States ex rel. Lawton v. Takeda Pharm. Co., 842 F.3d 125, 130 (1st Cir. 2016)(citations omitted).

[4] Sanderson v. HCA–The Healthcare Co., 447 F.3d 873, 877 (6th Cir.2006).

[5] 712 F.3d 21 (1st Cir. 2013), 712 F.3d 51 (1st Cir. 2013), and 712 F.3d 0 (1st Cir. 2013).

[6] Id. at 44-45.

[7] Case No. 16-1805, 2017 WL 395094 (1st Cir. January 30, 2017).

[8] Id. (emphasis in original).

[9] United States ex rel. Bogina III v. Medline Industries, Inc. et al., 809 F.3d 365, 367 (7th Cir. 2016).