Stats Without Facts: Another Court Rejects Data Mining Qui Tam Strategy
Jeff Bucholtz and Jeremy Bylund, King & Spalding LLP
The Ninth Circuit recently joined the Fifth Circuit in rejecting a data mining strategy employed in qui tam suits under the False Claims Act. These decisions are a welcome reminder that statistics are no substitute for facts showing fraud.
A Relator’s Statistics-Based Strategy
Traditionally, qui tam relators – plaintiffs who file False Claims Act lawsuits on behalf of the government – have been “insiders” like former employees who claim personal knowledge of alleged fraud. But the relator plaintiff in the Fifth Circuit and Ninth Circuit cases, Integra Med Analytics LLC, exemplifies a rising trend of “professional” relators—entities whose sole or primary business is to file qui tam actions.
Lacking any insider knowledge of actual fraud, Integra claimed in these cases to apply “proprietary statistical analysis” to Centers for Medicare & Medicaid Services (“CMS”) claims data to look for billing “outliers” among providers submitting claims to federal healthcare programs. Such “outliers,” according to Integra’s lawsuits, must be committing fraud.
Fortunately, two Circuits have now rejected this statistics-heavy, facts-light approach, holding that such an analysis is insufficient to plausibly plead fraud because it cannot rule out the obvious explanation that some healthcare providers are simply better at lawfully analyzing, understanding, and adapting to complex billing regulations and requirements.
In 2007, CMS – the federal agency, part of the Department of Health and Human Services, that administers the Medicare and Medicaid programs – reduced the base Medicare reimbursement rate for inpatient hospital care but increased the number of “secondary diagnosis” codes that could lead to higher reimbursement if certain complicating conditions or variables are present.
CMS encouraged hospitals to maximize their legal reimbursements under this change, stating, “we do not believe there is anything inappropriate, unethical or otherwise wrong with hospitals taking full advantage of coding opportunities to maximize Medicare payment that is supported by documentation in the medical record.” Medicare Program: Changes to the Hospital Inpatient Prospective Payment Systems and Fiscal Year 2008 Rates, 72 Fed. Reg. 47130, 47180 (Aug. 22, 2007). Hospitals had to adapt to this change in the reimbursement system and, as one would expect, some were faster than others at learning to “tak[e] full advantage of coding opportunities” to maximize appropriate payouts under the new system.
Integra filed qui tam actions against hospitals receiving higher reimbursement. Its complaints emphasized statistical analysis allegedly showing with near certainty that the significantly higher reimbursements to certain hospitals were not caused by random factors. Fortunately, two Circuits have rejected this approach.
Ninth Circuit Rejects Relator’s Strategy
Earlier this year, the Ninth Circuit reversed the district court’s denial of dismissal because “Integra offers only a possible explanation—that doctors lied about underlying medical conditions—to explain a statistical trend that is consistent with a plausible alternative (and legal) explanation.” Integra Med Analytics LLC v. Providence Health & Servs., __ F. App’x __, No. 19-56367, slip op. 9 (9th Cir. Mar. 31, 2021).
Integra’s claim failed to cross “the line between possibility and plausibility” because “Integra does not rule out an obvious alternative explanation[:] that Providence . . . was simply ahead of others in its industry.” Id. at 8–9. After all, CMS itself encouraged hospitals to maximize reimbursement, so the higher reimbursements could very well be indicative of lawful, “rational and competitive business strategy.” Id. at 9.
Fifth Circuit Rejects Relator’s Strategy
Last year, the Fifth Circuit likewise rejected a similar suit brought by Integra against another hospital because “statistical data cannot meet th[e] pleading requirements if, among other possible issues, it is also consistent with a legal and obvious alternative explanation.” United States ex rel. Integra Med Analytics, LLC v. Baylor Scott & White Health, 816 F. App’x 892, 898 (5th Cir. 2020).
Key Takeaways for Businesses
These cases demonstrate that statistical analysis showing outliers, “suspicious” patterns, and variations from larger trends should not get a relator past the pleading stage where there are obvious and innocent explanations that make an inference of fraud merely possible, not plausible. The cases may also suggest a healthy skepticism by the courts of relators that seek to turn the False Claims Act’s qui tam provisions into a lucrative business opportunity despite lacking actual knowledge of fraud.
Accusing someone of defrauding the government is a big deal, and False Claims Act cases impose massive costs on defendants as well as on the court system. It is only reasonable to insist that a relator plead actual facts showing fraud—not just questionable inferences from statistics—in order to unlock the doors of discovery.