Statistical Sampling As A Means to Prove Liability Continues to Gain Traction

Posted by Scott Stein and Brenna Jenny

A court in the Middle District of Florida is the latest of a growing number of courts (as reported here and here) that has allowed relators to rely on statistical sampling in order to establish liability in FCA cases involving large numbers of claims. See United States ex rel. Ruckh v. Genoa Healthcare, LLC, No. 11-cv-01303 (M.D. Fla. Apr. 28, 2015).

The relator, a former employee of the defendant operators of a chain of nursing and rehabilitation centers, initially filed a qui tam suit alleging upcoding at two facilities where she had worked. After the court dismissed the initial complaint for failing to include sufficient details about the alleged fraudulent scheme, the relator filed an amended complaint citing upcoding at fifty-three facilities.

Following denial of the motion to dismiss the amended complaint, and “[c]iting the voluminous discovery in this action and arguing that producing and processing the relevant medical records at the ‘fifty-three . . . [medical] facilities and some fifty-three . . . off-site storage locations’ within a reasonable time is impossible,” the relator filed a motion in limine for permission to submit an export report that would use statistical sampling and extrapolation in order to estimate the volume of overpayments allegedly retained by the defendants.

The United States filed a Statement of Interest in support of the relator’s motion (accessible here). The government opposed the defendants’ argument that the FCA necessarily requires proof of individual false claims, particularly in cases, such as this one, involving allegations of medically unnecessary care. While medical necessity decisions would be based on unique determinations for the population of patients in the sample, the government argued that so long as the sample was representative, the extrapolated result would be valid, and the defendants should be relegated to attacking the weight of the inferences through competing testimony.

The court relied heavily on the District of Tennessee’s decision in U.S. ex rel. Martin v. Life Care Centers of America, Inc. (discussed here) in ruling that the relator could use statistical sampling to estimate the claimed overpayments. In concluding that there is “no universal ban” on sampling in qui tam suits, court emphasized in particular the Martin court’s observation that proceeding without extrapolation “would consume an unacceptable portion of the Court’s limited resources.”

A copy of the court’s opinion can be found here.