As we discussed yesterday, the False Claims Act is often triggered by widespread, pervasive, long-term fraudulent behavior–resulting in voluminous records and hundreds of thousands of invoices evidencing the fraudulent misconduct. As a result, courts have begun to consider whether the use of a statistical random sample of admissions or patient files is an appropriate means by which the relator or federal government, or both, can establish liability on the part of a defendant. Understandably, defendants are loath to establish this standard, as using statistical samples will presumably make establishing False Claims Act liability easier and more efficient. By contrast, plaintiffs assert that the use of statistical sampling, when conducted correctly, paints a clear picture of fraud and misconduct while not overburdening the court with excessive and duplicative evidence.
In yesterday’s post, we introduced a recent case out of the U.S. District Court for the Eastern District of Tennessee wherein the federal government and a relator advanced the argument that statistical sampling is appropriate against an alleged healthcare fraud perpetrator. We continue today by exploring the court’s ultimate analysis in the case, including its concise review of whether the False Claims Act is an appropriate area of law for the use of statistical sampling.
Court decides whether statistical sampling belongs in False Claims Act litigation
The use of sampling is relatively new in the context of False Claims Act litigation, especially when used to establish actual liability. As discussed in yesterday’s post, the court first reviewed an extensive history of the use of sampling across several other areas of law, appearing encouraged by the historical pervasiveness of this tactic. From there, the court examined several False Claims Act cases where the court either allowed the sampling or it did not.
The court started its review with a federal case out of Massachusetts involving fraud against Medicare. The court in that case declined to employ statistical sampling, stating it preferred to review each claim individually. In that case, however, the government had presented just 676 claims of fraud–about 0.04% of the 154,000 claims present in the Martin case.
The court then reviewed cases permitting the use of statistical sampling, highlighting the routine use of this method in administrative law decisions, which the court ultimately concluded were not closely related to False Claims Act cases. The government was able to point to one more case wherein the court allowed statistical sampling to establish liability in a False Claims Act case. That case could still, however, be distinguished from the Martin case in that the court there relied on the sample after receiving no objection by the defendant. In essence, the sample was admitted by default.
Court weighs defendant’s arguments
The defendant’s main argument against the use of a sample is that each patient’s file is unique, containing “fact-intensive, subjective determinations made by scores of different physicians, therapist and other professionals as to whether individualized therapy treatments were medically necessary.” The defendant further asserted that the factors that are considered by a medical team when deciding if a patient needs therapy are far too intricate and establishing liability using a statistical sample would be impossible. The defendant provided a list of said factors, including “age; gender; pre-hospital condition/prior level of function; reason for hospitalization; condition upon admission to skilled nursing facility; visual defects; whether the patient received nursing assistance at home prior to hospitalization; incontinence; mental status; whether the medical history was extensive; whether the patient was terminally ill; whether the patient’s strength was impaired; degree of the patient’s dependency on others; whether the patient exhibited nausea, pain and fatigue, and endurance, among many others.”
Nonetheless, the court was not persuaded by the defendant’s argument. Specifically, the court held, “The large number of allegedly false claims at issue in this action leads to the natural effect that the claims are unique to one another in some respects.” The court found the defendant’s arguments to be speculative, siding with the government and its use of a qualified statistician to extrapolate an appropriate sample.
Last, the court applied a bit of its own extrapolation, recognizing the slippery slope that could ensue by disallowing the use of reliable statistical samples to establish False Claims Act liability. In its opinion, the court pointed out that parties engaging in massive, widespread, multi-jurisdictional, long-term acts of fraud would actually be protected by the notion that the plaintiff would need to present each and every invoice allegedly showing fraud. It opined, “If the Court were to reach the conclusion urged by the Defendant–that a claim-by-claim review is required in every FCA action and that statistical sampling is never permissible–potential perpetrators of fraud would be emboldened by the fact that a claim-by-claim review is often impractical.”
Aware of healthcare fraud? Contact us today.
If you are aware of fraud involving Medicare, Medicaid or other government healthcare enrollees, please contact Berger & Montague, P.C. today to discuss your options under the False Claims Act.