In recent years, the U.S. Department of Justice (“DOJ”) has increasingly leveraged data analytics to combat fraud. Principal Deputy Chief of DOJ’s Fraud Section, Joe Beemsterboer, described the department’s data-mining capabilities as the “foundation of how [DOJ] investigate[s] and analyze[s] cases,” and explained that digital forensics equips the department with “powerful” tools for identifying “trends,” “spikes” and “outliers.” Just over a year later, in December 2020, Michael Granston, Deputy Assistant Attorney General of DOJ’s Civil Division, echoed Mr. Beemsterboer’s remarks at the American Bar Association’s Civil False Claims Act (“FCA”) and Qui Tam Enforcement Institute. Mr. Granston previewed the “future” of DOJ’s FCA “enforcement efforts” by emphasizing that the public should “expect” DOJ to “expand its reliance” on “sophisticated analyses of Medicare data to uncover potential fraud schemes” by “identify[ing] trends and extreme outliers.” Mr. Granston went on to explain that Medicare data “can even allow [DOJ] to demonstrate and quantify sophisticated relationships” among alleged fraudsters.

These new investigative technology tools, however, are not exclusive to DOJ. Data mining, for example, also opens the door for opportunistic industry outsiders to formulate and pursue FCA liability theories against defendants with whom they have zero personal dealings. As the growing amount of publicly available data ushers in a new digital wave of fraud investigation and enforcement, members of the health care and life science industries have a pressing need for effective strategies to challenge such data-driven claims.

Importantly, in a little-noticed opinion issued last week, the United States Court of Appeals for the Ninth Circuit followed the lead of a decision issued last year by the Fifth Circuit confirming that defendants can defeat such claims at the pleadings stage. See Integra Med Analytics, LLC v. Providence Health & Servs., No. 19-56367 (9th Cir. Mar. 31, 2021); see also United States ex rel. Integra Med Analytics, L.L.C. v. Baylor Scott & White Health, 816 F. App’x 892 (5th Cir.), cert. denied, 141 S. Ct. 905 (2020).

Integra Med Analytics, LLC (“Integra”), a data forensic company and affiliate of the corporate whistleblower that successfully initiated one of the largest mortgage-backed securities fraud settlements, purports to leverage econometric and regression models to expose fraud, waste, and abuse in the health care industry. Relying on publicly available claims data obtained from the Centers for Medicare & Medicaid Services (“CMS”), Integra filed a series of FCA complaints against hospital systems and skilled nursing facilities across the country, including complaints against the Providence Health & Services system (“Providence”) in the Central District of California and the Baylor Scott & White Health system (“Baylor”) in the Western District of Texas. Integra alleged that Providence and Baylor, respectively, submitted over $250 million in false claims to Medicare by fraudulently claiming certain higher-value comorbidities. More to the point, Integra claimed its analysis of CMS’s public claims data identified Providence and Baylor as two of a handful of hospitals claiming various comorbidities significantly above the national average. Integra then alleged that it employed “unique algorithms and statistical processes” that purportedly ruled out demographic and other non-fraudulent reasons for this disparity and asserted that its CMS claims data analysis revealed a widespread “upcoding scheme.”

The Western District of Texas dismissed Integra’s lawsuit against Baylor with prejudice based on Baylor’s argument that Integra failed to state a plausible claim for relief under Rule 8 of the Federal Rules of Civil Procedure. The district court held that Integra failed to state a plausible claim because the alleged upcoding scheme was just as consistent with a lawful scheme to increase revenue through accurate documentation. On appeal, the Fifth Circuit affirmed Integra’s conclusory allegations were consistent with a legal and obvious alternative explanation, and the Supreme Court subsequently denied Integra’s petition for review.

Meanwhile, just a few weeks before the Western District of Texas’s opinion in Baylor, the Central District of California reached the opposite conclusion and denied Providence’s motion to dismiss Integra’s core FCA allegations on Rule 8 grounds. However, at Providence’s request, the district court later certified its order for interlocutory appeal to the Ninth Circuit, which accepted the appeal. After hearing oral argument earlier this year, the Ninth Circuit eventually reached the same conclusion as the Fifth Circuit, holding that Providence’s alleged conduct fell squarely in line with a lawful, rational, and competitive business strategy, and directing the district court to dismiss Integra’s complaint.

Data-mining relators now have both the resources (vast amounts of public data) and incentives (statutory right to share in FCA recoveries) to pursue sweeping fraud claims. DOJ likewise recognizes the utility of forensic data analysis in combating fraud and uses programmatic techniques to streamline investigations and expedite enforcement efforts. Therefore, members of the health care and life science industries should take note of the Providence and Baylor appellate decisions in light of the growing trend of algorithmic FCA investigations and litigation.

Disclosure: Reed Smith LLP served as counsel to the defendants in the Baylor case. The views expressed above are those of the author only.