The U.S. Department of Justice’s (the “Department”) Civil Division recently announced a new anti-fraud initiative dubbed the Fraud Oversight through Careful Use of Statistics (FOCUS) initiative. FOCUS is designed to formalize the Department’s engagement with so-called “data miner” relators who file qui tam complaints under the False Claims Act (FCA). The announcement signals the Department’s intent to both leverage a fast-growing source of FCA litigation and discourage data-miner relators from filing actions that cannot overcome a motion to dismiss. It carries significant implications for government contractors, grant recipients, healthcare providers, lenders, importers, and other recipients of federal funds whose dealings with the United States leave a public data trail.

The FOCUS Initiative

FOCUS is a response to the explosion of FCA qui tam lawsuits in recent years, culminating in a record 1,297 qui tam complaints in FY 2025. In FY 2026, there have already been 780 qui tam lawsuits filed—putting us on pace for another record-setting year. Much of the surge has been driven not by traditional insider whistleblowers (e.g., current and former employees), but by companies and individuals analyzing publicly available government data for potential signals of fraud.

Through FOCUS, the Civil Division seeks to provide guidance to would-be data-miner relators to ensure that their qui tam complaints are worthy of civil prosecutors’ time and have the potential of withstanding a motion to dismiss. Civil Fraud Section prosecutors will meet with data miners to assess their analytical capabilities, validation methods, and the reasons their data signals reliably correlate to actionable fraud. These pre-filing meetings are voluntary, not a precondition to filing. But the Department has signaled that it will preferentially work with data miners that demonstrate an investment in pre-filing diligence, analytical rigor, familiarity with program rules, and legally sufficient allegations. This may be another part of the Department’s broader interest in weeding out unmeritorious qui tam actions quickly, and potentially even seeking their dismissal.

Background: The Rise of the Data-Miner Relator

The traditional FCA relator is an industry insider: a former employee, contractor, vendor, or consultant with firsthand knowledge of alleged misconduct. The 1986 amendments to the FCA were structured around that paradigm, with bounty incentives of 15% to 30% of any recovery designed to encourage insiders to come forward. Over the last decade, however, a different model has emerged: “whistleblowers” who identify potential fraud solely through pattern analysis of publicly disclosed government data.

Data-miner relators typically rely on large federal data sets, including Medicare and Medicaid claims data published by the Centers for Medicare & Medicaid Services, Paycheck Protection Program (PPP) loan disclosures by the Small Business Administration, customs and trade data, and federal grant and contract award databases. The data miners apply statistical analyses, or, increasingly, machine-learning and AI tools, to flag anomalies or connections in the data that suggest fraud. PPP loans have been a particularly active driver of recent filings, with data miners cross-referencing SBA loan data against company filings, state corporate registries, news media, company websites, and other public sources to identify potential affiliation, headcount, or eligibility issues.

Mixed Results in the Courts

Data-miner relators have faced two structural defenses. First, the FCA’s public disclosure bar, 31 U.S.C. § 3730(e)(4), precludes qui tam suits substantially based on certain public disclosures unless the relator is an original source. Second, Federal Rule of Civil Procedure 9(b)’s requirement that fraud be pleaded with particularity, which applies to FCA claims, has proven difficult to satisfy for data miners, where statistical inference is their primary tool.

The leading appellate decisions reflect judicial skepticism of “stats without facts” pleadings. For example, the Ninth Circuit reversed the denial of a motion to dismiss where the relator, a self-described data analytics company with no insider knowledge, relied on statistical analyses of CMS claims data to allege a Medicare upcoding scheme.¹ The court held the data-miner relator had offered only a possible explanation for billing patterns that were equally consistent with a plausible, lawful alternative explanation, and therefore failed to cross the line from possibility to plausibility under Iqbal and Twombly. The Fifth Circuit reached a similar conclusion holding that statistical data cannot satisfy Rule 9(b) where it is equally consistent with a legal and obvious alternative explanation.²

The government has pushed back against opportunistic professional relators. In United States ex rel. CIMZNHCA, LLC v. UCB, Inc., the Department moved to dismiss a qui tam filed by an entity organized specifically for that qui tam action and by an organization that had established nearly a dozen other entities to pursue parallel Anti-Kickback Statute–based FCA cases against pharmaceutical manufacturers.³ The Department argued the cases lacked sufficient merit to justify the cost of investigation and were contrary to the public interest. The Seventh Circuit ultimately reversed the district court’s denial of the motion, observing that the FCA was not designed to operate as an investment vehicle for financial speculators.

But the caselaw cuts both ways. Where data miners have invested in genuine pre-filing investigation, supplemented public data with non-public information, or tied their analyses tightly to documented program requirements, courts have allowed suits to proceed past the pleading stage, and the Department has occasionally even intervened. PPP-related qui tam actions brought by data miner entities have produced significant recoveries and are principal drivers of the FY 2024 and FY 2025 surge in qui tam filings.

What the FOCUS Initiative Signals

Read against this backdrop, FOCUS reflects a few policy choices.

First, the Department is embracing the data-miner relator model. The Civil Division could have responded to the qui tam surge by pressing for more restrictive judicial constructions of the public disclosure bar or by expanding its use of dismissal authority under 31 U.S.C. § 3730(c)(2)(A). Instead, the Department is opening a formal channel for engagement, on the premise that the most sophisticated analytics partners may surface from the ocean of data frauds the Department or other whistleblowers might otherwise miss.

Second, the Department intends to triage and filter. By prioritizing data miners that can articulate validation methods, demonstrate pre-filing diligence, and show familiarity with program rules, the Department is implicitly distinguishing rigorous data miners from less sophisticated and more opportunistic ones. Cases built on raw statistical outliers without supporting facts, like the model the Fifth and Ninth Circuits have rejected, may be less likely to merit DOJ attention (or intervention). The Department has recently expressed increased interest in discouraging unmeritorious qui tam actions. FOCUS may be a significant component of that effort given the number of data-miner relator qui tam actions. Through FOCUS, DOJ cautions data-miner relators that they must: “provide valuable leads through high-quality, reliable, and predictive data analyses and signals and a thorough understanding of the relevant legal obligations”; “state with particularity the circumstances constituting fraud” under Rule 9(b); “assess potential alternative explanations for the observed conduct” and “articulate . . . both scienter and falsity”; and demonstrate “program eligibility requirements and relevant regulatory frameworks.”

Third, FOCUS complements the Department’s broader fraud-enforcement trends. With FCA recoveries setting a record at $6.8 billion in FY 2025, the cross-agency Trade Fraud Task Force ramping up tariff- and customs-related FCA enforcement, the Civil Cyber-Fraud Initiative continuing to mature, and DEI-related civil rights FCA initiatives recently announced, FOCUS adds yet one more effort toward the Department’s FCA enforcement push.

Considerations for FCA-Exposed Organizations

For organizations that directly or indirectly receive federal funds, news of the FOCUS Initiative reinforces the need to monitor developments in FCA enforcement and continue to invest in robust compliance programs. A few practical implications are worth highlighting:

Public data is potential enforcement data. Data an organization submits to or generates for the federal government, such as claims data, loan applications, customs entries, payroll certifications, grant certifications, and many others, may be or may soon be accessible to data-miner relators. That is a key risk vector, but also an opportunity, as organizations can pressure-test internal data against the same kinds of analyses third parties run externally, including outlier and anomaly detection on billing, pricing, and reporting data.

Documentation of legitimate variation could be a difference maker. The Fifth and Ninth Circuits’ rejections of data-miner relator FCA theories rested largely on the Courts’ willingness to credit lawful alternative explanations for statistical outliers. Organizations that can document why their billing, coding, payroll, or pricing patterns may differ from peers, such as because of case mix, geography, business model, regulatory interpretation, or other lawful drivers, may be substantially better positioned if a data miner’s complaint is unsealed or DOJ opens an investigation.

The value of pre-CID self-assessment and self-disclosure may be increasing. Once the Department engages with a data-miner relator through FOCUS, the path from analytic insight to civil investigative demand may be shorter than under the traditional insider-driven model. When an organization’s internal review surfaces genuine compliance issues, particularly in current Department priority areas, such as Medicare Advantage risk-adjustment coding, PPP eligibility and forgiveness, customs valuation and country-of-origin, and cybersecurity certifications, early voluntary disclosure and remediation could meaningfully affect resolution outcomes.