Thomas Prose v. Molina Healthcare of Illinois

In the United States Court of Appeals For the Seventh Circuit ____________________ No. 20-2243 UNITED STATES OF AMERICA and the STATE OF ILLINOIS ex rel. THOMAS PROSE, Plaintiff-Appellant, v. MOLINA HEALTHCARE OF ILLINOIS, INC., and MOLINA HEALTHCARE, INC., Defendants-Appellees. ____________________ Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 17 C 6638 — Virginia M. Kendall, Judge. ____________________ ARGUED JANUARY 15, 2021 — DECIDED AUGUST 19, 2021 ____________________ Before SYKES, Chief Judge, and WOOD and HAMILTON, Circuit Judges. WOOD, Circuit Judge. Sophisticated players in the healthcare market know that services come at a cost; providers charge fees commensurate with the services rendered; and payors expect to receive value for their money. There are 2 No. 20-2243 many options from which to choose when designing a payment scheme, including fee-for-service, prepaid services using the health-maintenance organization model (HMO), and capitation payments, to name just a few. Each of these models attempts to balance expected services against expected costs. The present case involves a capitation system, which is similar to the traditional HMO approach in which parties agree to a fixed per-patient fee that covers all services within the scope of a governing plan. Molina Healthcare of Illinois (Molina) contracted with the state’s Medicaid program (which in turn is largely funded by the federal government, see Illinois Medicaid, https://www.benefits.gov/benefit/1628) to provide multiple tiers of medical-service plans with scaled capitation rates. Among those, the Nursing Facility (NF) plan required Molina to provide Skilled Nursing Facility (SNF) services. Molina itself, however, did not deliver those services; instead, it subcontracted with GenMed to cover this obligation. Molina received a general capitation payment from the state, out of which it was to pay GenMed for the SNF component. But little time passed before Molina breached its contract with GenMed and GenMed terminated the contract. After GenMed quit, Molina continued to collect money from the state for the SNF services, but it was neither providing those services itself nor making them available through any third party. Molina never told the government about this breakdown, nor did it seek out a replacement service provider. Thomas Prose, the founder of GenMed, brought this qui tam action under both the federal and the state False Claims Acts. See 31 U.S.C. § 3729 et seq.; 740 ILCS 175/1 et seq. (Because the state law does not differ in any meaningful way from the federal law, we refer in this opinion only to the federal law for No. 20-2243 3 the sake of simplicity.) Prose alleged that Molina submitted fraudulent claims for payments to the Department (which was for the most part just a conduit for federal funds—a point we will not repeat) for skilled nursing facility services. Although the district court agreed with Prose that the SNF services were material to the contract, it dismissed the case at the pleading stage because it found that the complaint insuffi- ciently alleged that Molina knew that this condition was material. But on our independent reading of the complaint, we conclude that it plausibly alleges that as a sophisticated player in the medical-services industry, Molina was aware that these kinds of services play a material role in the delivery of Medicaid benefits. We therefore reverse and remand for further proceedings. I We present the facts in the light most favorable to Prose, the party opposing dismissal for failure to state a claim. Molina, a subsidiary of Molina Healthcare, Inc. (Molina Healthcare), is a Managed Care Organization (MCO). It has contracted with the Illinois Department of Healthcare and Family Services to provide healthcare services for Illinois Medicaid beneficiaries. Molina’s contract with the state was a “risk contract,” in which the parties agree to an expected cost for services for a patient and Molina assumed the risk that the cost of those services might exceed the contracted payment amount. 42 C.F.R. § 438.2 . As part of this risk contract, Molina and the Department agreed to capitation payments—periodic contractual fees, calculated per enrollee. These fees must be “actuarily sound.” Id. Each enrollment category had its own schedule of payments. A given category’s capitation rate reflected the anticipated 4 No. 20-2243 costs per person on an amortized basis. There was nothing unusual about this arrangement. In the late 1980s and 1990s, the capitation-payment model became common in the healthcare industry. It is similar to the more traditional health maintenance organization (HMO), in which a health insurance provider covers all care over a fixed annual fee, but it differs in some important ways. Capitation rates, in a word, are more flexible. They allow providers to establish distinct rate tiers, and the providers agree to delineate at the outset exactly what services they will furnish within each tier. Membership in each tier is correlated with factors such as age, health, and needed services. See, e.g., Nina Novak, Health Care Risk Contracting: The Capitation Alternative, 3 HEALTH LAW. 4, 4–5 (1987). A Managed Care Organization plays an active role in the creation of the plan, as it needs to understand the risk it is assuming through its guarantee of services. See Andrew Ruskin, Capitation: The Legal Implication of Using Capitation to Affect Physician Decision-Making Processes, 13 J. CONTEMP. HEALTH L. & POL’Y 391, 397, 409, 411 (1997). Molina’s contract created “rate cells” that were “stratified by age … , geographic services area (Greater Chicago and Central Illinois), and setting-of-care.” It defined five care settings: Nursing Facility, Waiver, Waiver Plus, Community, and Community Plus. The lowest cost and most populous of these cells was the Community group. For the Greater Chicago Community category during the contract period for February to December 2014, for example, the projected enrollee count was 261,108, and the monthly capitation rate the state paid to Molina was $53.51 for each person 65 years and older. By contrast, the highest-cost category—Nursing Facility—had 70,836 enrollees covered at a monthly rate of $3,180.30 per person 65 and older. Our case concerns this latter category. No. 20-2243 5 Molina contracted to provide Skilled Nursing Facility (universally abbreviated as SNF) services for Nursing Facility enrollees. Under Illinois state law, SNF providers, known as “SNFists,” are “medical professional[s] specializing in the care of individuals residing in nursing homes employed by or under contract with a MCO.” 305 ILCS 5/5F-15. Molina’s contract further specified that a SNFist’s “entire professional focus is the general medical care of individuals residing in a Nursing Facility and whose activities include Enrollee care oversight, communication with families, significant others, PCPs, and Nursing Facility administration.” SNFists perform valuable long-term care for sick, disabled, or elderly patients who need long-term medical and nursing care without hospitalization. Molina’s contract with the Department emphasized that SNFist services were integral to improving the enrollee’s quality of life and potentially to enabling her to be discharged from the nursing home. 1 In order to deliver these expensive services, in April 2014 Molina entered into an agreement with GenMed, because Molina did not have the necessary qualified personnel. This contract provided that GenMed would provide SNF services for 1 The dissent suggests that the SNFist services provided by Molina were contractually limited to “care coordination and management.” That was true in some circumstances, but not all. Providers employed through the SNFist program were also expected to “deliver care” “when appropriate or necessary.” And in its general definition of SNF facility services, the contract included all of “Skilled Nursing care, continuous Skilled Nursing observations, restorative nursing, and other services under professional direction with frequent medical supervision.” Construing the allegations in the light most favorable to Prose, as we must, this shows that SNFist services are comprehensive, not just one of a bundle of 20 or 30 different items, as the dissent contends. 6 No. 20-2243 Molina’s Nursing Facility enrollees. The Department was not a party to the contract, and so it continued to pay Molina the full capitation payments for the SNF recipients. Molina then used those funds to pay GenMed the agreed amount. This arrangement, however, lasted only about nine months. In January 2015, Molina stopped reimbursing GenMed and sought to renegotiate the price terms of the service agreement. GenMed continued to provide SNF services through March 2015, but it terminated the contract on April 2, 2015, after Molina continued to refuse to pay it. From April 2, 2015, until at least April 5, 2017, Molina was not delivering SNF services to anyone, either with its own personnel or through a subcontractor. Indeed, it was not even looking for a replacement for GenMed. It did not inform the Department or the federal authorities of this change, and so the Department continued to pay it the full capitation amount for SNF services—in essence, payments for nothing. Aware of the situation because of his association with GenMed, Thomas Prose filed this qui tam action on September 14, 2017, alleging that Molina violated the False Claims Act by seeking and obtaining compensation despite failing to provide material services under its contract with the Department. II Since we are evaluating the district court’s decision to grant a motion to dismiss under Rule 12(b)(6), we accept all well-pleaded facts as true and draw all reasonable inferences in favor of the non-moving party. O’Brien v. Village of Lincolnshire, 955 F.3d 616 , 621 (7th Cir. 2020). Critically, however, this is not a case that is governed by the usual notice-pleading standards of Federal Rule of Civil Procedure 8. See, e.g., Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). A party bringing a case No. 20-2243 7 alleging fraud must satisfy the heightened pleading standards set forth in Rule 9(b), which says that “[i]n alleging fraud or mistake, a party must state with particularity the circumstances constituting fraud or mistake.” FED. R. CIV. P. 9(b). At the same time, Rule 9(b) carves out several matters that may be alleged generally, including “[m]alice, intent, knowledge, and other conditions of a person’s mind.” Id. Rule 9(b)’s more demanding pleading requirements apply to suits brought under the False Claims Act. The complaint must describe the “who, what, when, where, and how” of the fraud to survive a motion to dismiss. United States ex rel. Presser v. Acacia Mental Health Clinic, LLC, 836 F.3d 770, 776 (7th Cir. 2016) (quoting United States ex rel. Lusby v. Rolls– Royce Corp., 570 F.3d 849, 853 (7th Cir. 2009)). Nonetheless, courts and litigants should not “take an overly rigid view of the formulation”; the allegation must be “precis[e]” and “substantiat[ed],” but the specific details that are needed to support a plausible claim of fraud will depend on the facts of the case. Presser, 836 F.3d at 766 (quoting Pirelli Armstrong Tire Corp. Retiree Med. Benefits Tr. v. Walgreen Co., 631 F.3d 436, 442 (7th Cir. 2011)). As we noted earlier, the Illinois False Claims Act applies the same standards as the federal statute. Bellevue v. Universal Health Servs. of Hartgrove, Inc., 867 F.3d 712 , 716 n. 2 (7th Cir. 2017). A Before assessing Prose’s complaint, it is helpful to take a more detailed look at the False Claims Act. This statute creates a right of action under which private parties may, on behalf of the federal government, bring lawsuits alleging fraud. 31 U.S.C. § 3730 (b). The actions go by the hoary Latin term “qui tam” (short for qui tam pro domino rege quam pro se ipso in 8 No. 20-2243 hac parte sequitur, meaning “who as well for the king as for himself sues in this matter,” see Bryan A. Garner, ed., BLACK’S LAW DICTIONARY at 1444 (10th ed. 2014)). The party seeking to represent the government’s interest is called a “relator.” Successful relators are motivated by the prospect of recovering sizable shares of the money paid to the government after bringing a successful claim. Glaser v. Wound Care Consultants, Inc., 570 F.3d 907, 912 (7th Cir. 2009). The government has the right, but is not obligated, to proceed on a claim brought by a relator; it may elect to dismiss the action notwithstanding the party’s objection. 31 U.S.C. § 3730 (c)(2)(B). When the government chooses not to proceed with the action but does not dismiss the action either, the initiating party retains the right to proceed against the defendant. 31 U.S.C. § 3730 (c)(3). The Act makes it unlawful knowingly (1) to present or cause to be presented a false or fraudulent claim for payment to the United States, (2) to make or use a false record or statement material to a false or fraudulent claim, or (3) to use a false record or statement to conceal or decrease an obligation to pay money to the United States. United States ex rel. Yannacopoulos v. Gen. Dynamics, 652 F.3d 818, 822 (7th Cir. 2011). A successful claim requires proof both that the defendant made a statement to receive money from the government and that he made that statement knowing it was false. Id. But there is more. Not all false statements are actionable under the Act. The plaintiff also must prove that the violation proximately caused the alleged injury. United States v. Luce, 873 F.3d 999 , 1011–14 (7th Cir. 2017). In other words, the pecuniary losses must be “within the foreseeable risk of harm” that the false statement created. Id. at 1012 . In addition, the defendant’s conduct must meet a strict materiality requirement. Universal Health Servs., Inc. v. United States ex rel. Escobar, 136 S. Ct. 1989 , No. 20-2243 9 1996 (2016). It is not enough simply to say that the government required compliance with a certain condition for payment. The facts must indicate that the government actually attaches weight to that requirement and relies on compliance with it. In sum, as the Third Circuit has put it, the relator must establish (1) falsity, (2) causation, (3) knowledge, and (4) materiality. United States ex rel. Petratos v. Genentech Inc., 855 F.3d 481, 487 (3d Cir. 2017). The Act is not limited to claims that are facially false. It covers a defendant’s more general decision fraudulently to procure payment from the government. Consequently, while the archetypical claim is one in which a “claim for payment is itself literally false and fraudulent,” United States ex rel. Hendow v. Univ. of Phoenix, 461 F.3d 1166, 1170 (9th Cir. 2006), courts have identified particular theories that support FCA claims, including (1) false certification to the government that the party has complied with a statute, regulation, or condition of payment; (2) promissory fraud, or fraud in the inducement, id. at 1172–73; and (3) implied false certification, see Escobar, 136 S. Ct. at 2001 . The implied false certification claim involves the omission of key facts rather than affirmative misrepresentations. This type of liability arises if the “defendant makes representations in submitting a claim but omits its violations of statutory, regulatory, or contractual requirements[;] those omissions can be a basis for liability if they render the defendant’s representations misleading with respect to the goods or services provided.” Id. at 1999 . B Prose’s complaint raises allegations under all three of these approaches: factual falsity, fraud in the inducement, and implied false certification. At the same time, he contends 10 No. 20-2243 that these labels should be jettisoned. Taking our guidance from Escobar, we decline to distill one unified approach for all cases. The Court’s focus on the implied false certification theory in Escobar signals that it continues to find that there are distinct ways in which the statute may be violated. We will follow suit. As we now explain, we conclude that Prose has adequately stated a claim under the Act. His detailed allegations support a strong inference that Molina was making false claims. At this stage, that is enough; as the litigation proceeds, it is possible that one or more of these theories will lack support. But there is time enough for that assessment at trial or upon a motion for summary judgment. Fraud is a serious matter. Rule 9 represents a policy decision to protect potential fraud defendants from litigation based on nothing but rumor or speculation. Instead, the relator must set forth the basis for her conclusion that fraud is afoot. United States ex rel. Grenadyor v. Ukrainian Vill. Pharmacy, Inc., 772 F.3d 1102, 1108 (7th Cir. 2014). But as we have been saying, that does not require the impossible. Relators with a legitimate basis for bringing False Claims Act cases will not generally have propriety information of the company they are trying to sue, and so courts do not demand voluminous documentation substantiating fraud at the pleading stage. All that is necessary are sufficiently detailed allegations. We begin with the allegations that would support a claim for direct factual falsity—the canonical FCA claim. The question is whether Prose’s allegations alert Molina in sufficient detail for Rule 9(b) purposes of how it allegedly made a “claim for payment [that] is itself literally false and fraudulent.” Hendow, 461 F.3d at 1170 . Prose contends that after April No. 20-2243 11 2, 2015, Molina submitted to the government materially fraudulent enrollment forms for each new enrollee in the Nursing Facility category of patients. As of that date, its contract with GenMed had ended, and it could not, and did not, provide SNF services. Rule 9(b) requires specificity, but it does not insist that a plaintiff literally prove his case in the complaint. Prose provided numerous details indicating when, where, how, and to whom allegedly false representations were made. He hardly can be blamed for not having information that exists only in Molina’s files. He did provide information that plausibly supports the inference that Molina included false information about the pertinent services for new enrollees. How else could it have asked for its capitation payments based on these additional beneficiaries? A direct assertion that Molina had new enrollees who were in the skilled nursing facility tier, coupled with an assertion that Molina was seeking reimbursement for their SNF services, is not an omission. It is a statement, and in this case a statement that Prose asserts was false. He did not need any more to defeat the challenge to the adequacy of his complaint. Prose also alleged circumstantial evidence of promissory fraud, or fraud in the inducement. Here, he needed to alert Molina with the necessary specificity of how it allegedly misrepresented its compliance with a condition of payment in order to induce the government to enter into a contract. Hendow, 461 F.3d at 1172–73 (citing United States ex rel. Marcus v. Hess, 317 U.S. 537, 542 (1943)); cf. United States v. Sanford-Brown, Ltd., 788 F.3d 696, 709 (7th Cir. 2015), abrogated by Escobar, 136 S. Ct. 1989 (“[F]raud entails making a false representation, such as a statement that the speaker will do something it plans not to 12 No. 20-2243 do.”). Prose charges that Molina fraudulently induced the Department to enter into contract renewals with Molina in 2016 and 2017 by affirmatively misrepresenting that it would continue to provide SNF services in its package for NF-category enrollees while not intending to do so. The district court concluded that the complaint in this respect fell short because it did not include any details about the contract-renewal negotiations between Molina and the Department. But how would Prose have had access to those documents or conversations? The obligation to set out the “who, what, when, where, and how” of the fraud does not require such granular detail. Prose set forth precise allegations about the beneficiaries, the time period, the mechanism for the fraud, and the financial consequences. Once again, at trial or upon a motion for summary judgment he will face a different burden, but for now, this was enough. Claims of fraudulent inducement also require the plaintiff to show that the defendant never intended to perform the promised act that induced the government to enter the contract. United States ex rel. Main v. Oakland City Univ., 426 F.3d 914, 917 (7th Cir. 2005) (“[F]ailure to honor one’s promise is (just) breach of contract, but making a promise that one intends not to keep is fraud.”). Prose put Molina on notice of this aspect of his case, too. He included details about statements made by Molina’s chief operating officer (COO), Benjamin Schoen, who stated in his deposition that Molina’s “staff did not have the ability or licensure to render [SNF] services.” Taken together with Molina’s defunct contract with GenMed and its failure to seek out a replacement SNF provider, the complaint alleges that any promise by Molina to No. 20-2243 13 provide SNF services during the contract-renewal process was fraudulent on its face. This may even have been more detail than was necessary, taking into account the fact that Rule 9(b) permits intent to be alleged generally. Construing the allegations liberally, the complaint asserts that Molina made some representations about actual SNF services that would be offered. Schoen acknowledged that Molina did not have the personnel available to perform those services. The complaint thus concludes that Molina did not and never intended to seek out another SNF service provider. This sufficed to allege intent. Finally, even if the complaint fell short of the required specificity under Rule 9 for the first two approaches, it was sufficient to state a claim for implied false certification. The Supreme Court described that version of fraud as follows in Escobar: … [T]he implied false certification theory can be a basis for liability. Specifically, liability can attach when the defendant submits a claim for payment that makes specific representations about the goods or services provided, but knowingly fails to disclose the defendant’s noncompliance with a statutory, regulatory, or contractual requirement. In these circumstances, liability may attach if the omission renders those representations misleading. We further hold that False Claims Act liability for failing to disclose violations of legal requirements does not turn upon whether those requirements were expressly designated as conditions of payment. Defendants can be liable for violating requirements even if 14 No. 20-2243 they were not expressly designated as conditions of payment. Conversely, even when a requirement is expressly designated a condition of payment, not every violation of such a requirement gives rise to liability. What matters is not the label the Government attaches to a requirement, but whether the defendant knowingly violated a requirement that the defendant knows is material to the Government’s payment decision. 136 S. Ct. at 1995–96. Even before Escobar, courts recognized express false certification—that is, an affirmative misstatement of compliance with a statute, regulation, or other contractual obligation to obtain payment from the government—as a basis of liability. United States ex rel. Absher v. Momence Meadows Nursing Ctr., Inc., 764 F.3d 699 , 710–11 (7th Cir. 2014). Implied and express statements raise distinct issues, however, and so Molina is mistaken when it suggests that the implied version requires express representations about the goods or services to be provided. Material omissions can suffice. Implied false certification is just another genre of fraud, and so plaintiffs must as usual satisfy Rule 9(b)’s requirements to plead falsity, materiality, and causation with particularity. (Knowledge is also an element, but it falls within the Rule’s carve-out.) As the Supreme Court did in Escobar, we focus first on the “rigorous materiality requirement” that the plaintiff must meet. 136 S. Ct. at 1996 . A misrepresentation is not material “merely because the Government designates compliance with a particular statutory, regulatory, or contractual requirement as a condition of payment.” Id. at 2003 . Such a stipulation is “relevant, but not automatically dispositive.” Id. But materiality requires more: typically, proof either that No. 20-2243 15 (1) a reasonable person would view the condition as important to a “choice of action in the transaction” or (2) the defendant knew or had reason to know that the recipient of the representation attaches importance to that condition. Id. at 2002–03. Should the government decide to pay despite knowing of the party’s noncompliance, that would be “very strong evidence” (though not dispositive) that the condition is not material. Id. In short, facts matter. The complaint must include allegations that show that the omission significantly affected the government’s actions. Prose’s complaint points to many factual representations that Molina made that, he charges, amounted to implied false certification. For instance, he alleges that Molina’s contract with the Department carefully created different rate cells for enrollees based on the level of care they would need; the level of care in turn yields a reasonable estimate of cost for each tier. Both are essential if the capitation payments are to be actuarially sound. The difference between the Community group and the Nursing Facility group is a whopping $3,127 per head. The middle-tier group costs roughly $600 less apiece than the Nursing Facility group. The size of the price differential alone offers strong support for a finding of materiality: it is hard to see why the government would be indifferent about paying $3,180 for services that should have been at the $54 level. The district court concluded that each enrollment form, which constituted a specific request for payment connected to the NF enrollees, was “impliedly false because it requested payment of the SNF capitation rate” when those services were not being rendered. 2 2 The dissent takes issue with the numbers here, asserting that Molina provided “something close to high-tier service at high-tier rates.” But that 16 No. 20-2243 Molina responds that the enrollment forms did not contain misleading omissions because Molina did not fraudulently manipulate the beneficiary pool to increase the number of people in the more expensive category. But that is just one way in which liability could be shown; it is not the only one. The complaint, read in Prose’s favor, contains specific allegations showing that Molina was far from a passive recipient of a favorable capitation rate. Prose was not relying on Molina’s receipt of capitation payments for existing enrollees. Rather, the complaint alleges that by submitting enrollment forms for new enrollees after Molina canceled its contract with GenMed, Molina implicitly falsely certified that Nursing Facility enrollees had access to SNF services. But they did not. Construed in Prose’s favor, the complaint describes Molina’s noncompliance with a contractual requirement to provide SNF services to Nursing Facility enrollees. This is akin to the defendant’s actions in Escobar, in which the Court found that the defendant “misleadingly omit[ted] [the] critical facts” that its care providers were not qualified to render services for which it nevertheless requested payments. 136 S. Ct. at 2001 . Molina’s strongest argument against materiality relies on its contention that the government continued to contract with Molina after learning that Molina could no longer provide SNF services. Molina emphasized that the government not claim appears to spring from the redefinition of SNFist services as nothing more than care coordination—a definition that neither the contract nor the pleadings reflect. Just how close to “high-tier services” Molina got is best decided on summary judgment or at trial, not here. For present purposes, Prose’s complaint adequately alleges that SNFist services explain much of the cost difference between the Nursing Facility tier and the less expensive tiers. No. 20-2243 17 only continued paying it after Prose brought this case, but it also renewed its contract with Molina twice during that time. It is true that the government’s continued payment of a claim despite “actual knowledge” that certain requirements are not met “is very strong evidence that those requirements are not material.” Escobar, 136 S. Ct. at 2003 . But this argument is better saved for a later stage, once both sides have conducted discovery. At this juncture, it appears that Molina is offering only part of the story. Later exploration will be needed before anyone can say what the government did and did not know about Molina’s provision of SNF services. For pleading purposes, Molina’s barebones assertion that the government was aware of all material facts is not enough to sweep away the elaborate facts that Prose furnished. The contract itself, which fixes the cost of the NF category well above the other tiers, is powerful evidence of the materiality of the SNF services. See Ruckh v. Salus Rehab., LLC, 963 F.3d 1089 , 1105 (11th Cir. 2020) (finding materiality when the issue “went to the heart” of the bargain). Many things could explain the government’s continued contracting with Molina. It may have expected to purge the underserved NF enrollees from the books; it may have needed time to work out a way not to prejudice Medicaid recipients who had nothing to do with this problem. Medicaid (along with the Children’s Health Insurance Program, or CHIP) serves more than 71 million people nationally and accounts for $600 billion in federal spending. See Center for Medicare and Medicaid Services, Medicaid Facts and Figures, at https://www.cms.gov/newsroom/factsheets/medicaid-facts-and-figure. An organization like that does not turn on a dime. 18 No. 20-2243 For all these reasons, we conclude that Prose’s complaint adequately alleged materiality for purposes of his qui tam action. The district court was also willing to go that far. Where Prose foundered, it thought, was on the final element of the claim: knowledge. The court found that the complaint failed adequately to allege that Molina knew that the government viewed SNF services as material. In Escobar, the Supreme Court identified a two-layered knowledge requirement: the defendant must (1) knowingly violate a requirement while (2) knowing that the government viewed the requirement as material to payment. 136 S. Ct. at 1996 . Even though Molina necessarily knew that it had violated the contractual requirement to provide SNF services, the district court thought that Prose’s allegations that Molina knew that these services were material were conclusory and need not be accepted as true. The allegations, it said, at most supported a conclusion that Molina’s actuarial consultants coordinated the payment scheme with the government. Missing, it thought, was a contention that Molina was involved in calculating the capitation rates. This was error. First, the court failed to give proper weight to the complaint’s description of Molina as a highly sophisticated member of the medical-services industry. Molina was quite familiar with capitation rates, and it knew that they are designed to allow the provider to be reimbursed for services rendered. And recall that this was a risk contract: Molina had a strong incentive to ensure that the capitation rate was high enough to cover its costs plus a reasonable profit, because it would be left holding the bag if the rate were too low. Ruskin, supra, at 397, 409; Novak, supra, at 5. In addition, knowledge may be alleged generally, even in a case under Rule 9(b), and so the district court was wrong to No. 20-2243 19 insist that Prose identify concrete evidence of actual knowledge. A party seeking to establish liability under the FCA may satisfy the Act’s knowledge requirement through proof of actual knowledge, deliberate ignorance of truth or falsity, or reckless disregard for truth or falsity. 31 U.S.C. 3729(b)(1)(A)(i)–(iii). Construing the allegations in Prose’s favor, there is ample detail to support a finding that Molina either had actual knowledge that the government would view skilled nursing services as a critical part of the Nursing Facility rate cell (i.e., as material), or that it was deliberately ignorant on this point. Once again, these high-cost services, essential to the nursing-home population, were the very reason why the government paid a capitation rate more than fifty times as much to support them. Molina subcontracted for SNF services because it could not provide those services. Its contract with GenMed recognized that these SNF services “fill the primary care gap” for Nursing Facility patients. The deal fell apart when Molina attempted to renegotiate its contract with GenMed to reduce the cost of those services and thus to increase its own profit margin. Molina therefore knew these services’ cost and their importance, and it knew that it was unable to provide these services without a partner such as GenMed. Prose’s complaint plausibly alleges this knowledge, insofar as it notes that before the actuarial consultant’s resolution of the cost breakdown, Molina and the government discussed these services at the proposal stage. Requiring more concrete proof of knowledge would run afoul of Rule 9(b). In light of this, we need not rely on Prose’s other arguments. He alleges a scheme to cover up Molina’s noncompliance by having its own personnel perform non-skilled work 20 No. 20-2243 for the nursing, such as face-to-face comprehensive assessments and annual comprehensive exams. That practice does not shed much light on the problem: Molina always admitted that its personnel were not qualified to provide SNF services, and it appears that these exams were merely non-SNF functions that Molina had delegated to GenMed. Last, we say a word about causation. This too is an element of an FCA claim: the plaintiff must establish that the defendant’s fraud “was a material element and a substantial factor in bringing about the injury.” Luce, 873 F.3d at 1012 (internal quotation omitted). Causation here is evident. By submitting enrollment forms requesting payment for services Molina could not provide to Illinois Medicaid, Molina caused the government to pay significant sums that it would not have paid with full knowledge. That is enough to satisfy the pleading burden on causation. Prose’s complaint sufficiently alleges that Molina knew that SNF services played a major role in the significantly higher capitation rate for the NF category. It thus suffices for purposes of his False Claims Act theories. We of course express no opinion on the ultimate fate of this litigation; we hold only that Prose may proceed. III The final loose end we must address is Molina Healthcare’s request that it be dismissed from the case. Molina Healthcare (as we briefly noted at the outset) is Molina’s parent company. It contends that corporate affiliation is not enough to support its liability. Given the decision on the merits, the district court did not reach the question of parent-company liability. Neither do we; it is far No. 20-2243 21 too underdeveloped at this point. But it is an issue that, if properly raised again, the district court should address on remand. The judgment of the district court is REVERSED and the case REMANDED for further proceedings consistent with this opinion. 22 No. 20-2243 SYKES, Chief Judge, dissenting. “The False Claims Act is not ‘an all-purpose antifraud statute[]’ or a vehicle for punishing garden-variety breaches of contract or regulatory violations.” Universal Health Servs., Inc. v. United States ex rel. Escobar, 136 S. Ct. 1989, 2003 (2016) (quoting Allison Engine Co. v. United States ex rel. Sanders, 553 U.S. 662, 672 (2008)). Our own precedent aligns with this understanding of the FCA’s reach. See United States v. Sanford-Brown, Ltd. (“Sanford-Brown II”), 840 F.3d 445, 447 (7th Cir. 2016). The majority moves our circuit law in a different direction, establishing a new rule that a mere request for payment from the government, coupled with material noncompliance with a contractual condition, is a cognizable FCA violation subject to the full panoply of remedies authorized by the Act, including qui tam suits and treble damages. Because that rule conflicts with Escobar and circuit precedent, I respectfully dissent. * * * The government and Molina Healthcare of Illinois have a risk contract. Each month the government pays Molina a fixed sum to provide health coverage for a Medicaid beneficiary, and no matter how expensive that beneficiary’s medical costs are, Molina is responsible. Molina profits when the fixed sum—the “capitation rate”—exceeds actual expenses; it swallows a loss when expenses are in excess. The contract creates five risk pools called “rate cells” that correspond to health status, and it fixes capitation payments by rate cell—higher capitation rates are paid for rate cells that are likely to require more intensive care. The most expensive of these rate cells is for an enrollee living in a nursing facility. To enroll a beneficiary, Molina submits a No. 20-2243 23 form to the government categorizing the enrollee by rate cell, and in response the government pays Molina the corresponding amount. Molina’s contract with the government specifies the “covered services” that it must provide to enrollees depending on their rate cells. As relevant here, an enrollee who resides in a skilled nursing facility is entitled to “SNFist” services, generally described as “intensive clinical management of Enrollees in Nursing Facilities.” Plaintiff–relator Thomas Prose alleges that for approximately two years, Molina submitted enrollment forms to the government but knowingly did not deliver SNFist services to its nursingfacility enrollees. To place this allegation in proper context, some background on the nature of these services is needed. The term “SNFist” is defined in the contract as a medical professional “whose entire professional focus is the general medical care of individuals residing in a Nursing Facility and whose activities include Enrollee care oversight, communication with families, significant others, [primary-care providers], and Nursing Facility administration.” Or as the contract puts it more succinctly, a SNFist is a medical professional who “provide[s] Care Management and care coordination activities” for enrollees residing in nursing facilities. Importantly, “care management” is not the direct provision of medical care, personal care, or social services to nursing-home residents; rather, as the contract defines the term, “care management” comprises “[s]ervices that assist Enrollees in 24 No. 20-2243 gaining access to needed services, including medical, social, education, and other services.” 1 (Emphasis added.) The contract gives Molina the option to provide SNFist services “either through direct employment or a subcontractual relationship,” and its “SNFist Program” may use either a “facility-based Provider (Physician or nurse practitioner)” or “telephonic or field-based Registered Nurses or licensed clinical social workers,” depending on the circumstances. Because SNFist services are provided only to enrollees in nursing facilities, it’s reasonable to assume that the inclusion of these services plays at least some role in the difference between the capitation rate for the nursing-facility rate cell and the rate cell below it. How large a role is unclear; a key question is whether Prose has alleged sufficient facts to show that the delivery of SNFist services was material to the government’s decision to pay Molina for nursing-facility enrollees during the relevant time period. I will return to the materiality point later. For now, it’s enough to note that SNFist services are one component of nursing-home care among many, and as explained, are contractually defined as care coordination and management. Moreover, a nursinghome enrollee is inherently a riskier beneficiary for Molina 1 The majority uses the term “SNF services,” which loosely suggests that what’s at stake here is a broader spectrum of nursing-facility services. Not so. Prose’s complaint alleges that from April 2, 2015, to April 5, 2017, Molina failed to deliver “SNFist services,” a contractually defined term that is limited to care coordination performed by a SNFist—not a broader set of services provided by skilled nursing facilities. No. 20-2243 25 to cover than a lower-tier enrollee, which also partly explains the difference in capitation rates. In the majority’s view, because Prose has alleged that Molina billed the government for the full nursing-facility capitation rate while failing to provide SNFist services, he has adequately pleaded an FCA claim for making materially false statements to the government. That reasoning might have surface appeal, but once we understand that SNFist services are just one component of nursing-home care among many, the error in the majority’s reasoning becomes clear. Prose’s complaint states a claim for breach of contract, but it relies on too many factually unsupported inferences to state a claim for an FCA violation. * * * My colleagues begin the analysis by identifying the three recognized theories of FCA liability: fraud in the inducement (or promissory fraud), express factual falsity, and implied false certification. Majority op. at 9. They also explain that Rule 9(b) of the Federal Rules of Civil Procedure requires an FCA plaintiff to plead fraud allegations with particularity rather than simply satisfy the usual plausibility standard. Id. at 10. I have no disagreement with these basic doctrinal points. The majority concludes, however, that even under Rule 9(b)’s demanding standards, Prose has stated an FCA claim under all three theories. In my view the complaint does not satisfy the heightened pleading standard under any of these theories. A. Fraud in the Inducement Prose alleges that Molina fraudulently induced the government to renew its contract in 2016 and 2017 by represent- 26 No. 20-2243 ing that it would provide SNFist services for nursing-facility enrollees while never intending to do so. I agree with the district judge that Prose’s allegations are too generalized and conclusory to state a claim under this theory. To satisfy Rule 9(b), “[t]he complaint must state the identity of the person making the misrepresentation, the time, place, and content of the misrepresentation, and the method by which the misrepresentation was communicated.” United States ex rel. Grenadyor v. Ukrainian Vill. Pharmacy, Inc., 772 F.3d 1102, 1106 (7th Cir. 2014) (quotation marks omitted). Prose’s complaint falls far short of checking these boxes. It includes no details of the contract renewals in 2016 and 2017 and does not point to any specific misleading statement made by an identified Molina representative, let alone specify the “time, place, and content” of the statement. The allegations of promissory fraud are not only vague and highly generalized, but they are made “[o]n information and belief,” which is insufficient under Rule 9(b). United States ex rel. Bogina v. Medline Indus., Inc., 809 F.3d 365, 370 (7th Cir. 2016) (“‘[O]n information and belief’ can mean as little as ‘rumor has it that … .’”). In essence, Prose simply invites us to assume that because the contract was renewed at a time when Molina was not providing SNFist services, Molina necessarily made false statements to the government. Surprisingly, the majority accepts this invitation to deviate from Rule 9(b) and forgives Prose for not describing the “who, what, when, where, and how” of the fraud, as required by the rule. United States ex rel. Presser v. Acacia Mental Health Clinic, LLC, 836 F.3d 770, 776 (7th Cir. 2016) (quotation marks omitted). My colleagues say that Prose cannot be expected to provide these factual particulars at the No. 20-2243 27 pleading stage because he lacks access to information about the contract-renewal discussions until discovery opens that door. Majority op. at 10–11. But we are not at liberty to loosen pleading standards under circumstances where a specific false statement is hard to identify. Rule 9(b) raises the pleading burden “because of the stigmatic injury that potentially results from allegations of fraud.” Presser, 836 F.3d at 776 . Pleading a fraud claim is challenging, but that’s the point: the rule “deters the filing of suits solely for discovery purposes” and “guards against the institution of a fraud-based action in order to discover whether unknown wrongs actually have occurred.” 5A ARTHUR R. MILLER ET AL., FEDERAL PRACTICE & PROCEDURE § 1296 (4th ed. 2021). By permitting Prose to proceed on generic allegations of promissory fraud pleaded “on information and belief,” this case will become the very “fishing expedition” that Rule 9(b) is meant to avoid. Vicom, Inc. v. Harbridge Merch. Servs., Inc., 20 F.3d 771, 777 (7th Cir. 1994). B. Express Factual Falsity As my colleagues explain, the archetypal FCA violation is an express factual falsehood—a “claim for payment [that] is itself literally false or fraudulent.” United States ex rel. Hendow v. Univ. of Phoenix, 461 F.3d 1166, 1170 (9th Cir. 2006). The majority reasons that Molina’s enrollment forms were factually false on their face because they amounted to “[a] direct assertion that Molina had new enrollees who were in the skilled nursing facility tier, coupled with an assertion that [it] was seeking reimbursement for their SNF services.” Majority op. at 11. This reasoning extends the factual-falsity theory too far. 28 No. 20-2243 A direct falsehood is an affirmative misrepresentation, not an omission. For example, in Presser the plaintiff alleged that a medical clinic submitted claims to the government for payment using billing codes corresponding to specific psychiatric services but in fact had performed only nonpsychiatric evaluations. 836 F.3d at 778–79. Thus, the clinic made an affirmative factual misrepresentation: it billed the government specifically for service X when it actually provided service Y. Id. at 779 (“Acacia … allegedly billed Medicaid for a completely different treatment. The claim therefore does not involve a misrepresentation by omission; it involves an express false statement.”). Here, by contrast, Prose alleges a falsehood by omission: Molina requested capitation payments at the nursing-facility rate without disclosing that it did not deliver one of the many services required by the contract. This allegation does not describe an affirmative misrepresentation. At most, it alleges a fraudulent omission, which situates this case within the theory of implied false certification. We should analyze Prose’s complaint under that framework, not expand the theory of facial factual falsity to include misleading omissions. That was the approach taken by the Supreme Court in Escobar. There, the plaintiffs alleged that a medical-services contractor submitted claims for payment to the government for counseling services it had provided and listed billing codes and identification numbers corresponding to the specific services its counselors had provided, along with their job titles, respectively. The problem was that the counselors “lacked licenses to provide mental health services, yet—despite regulatory requirements to the contrary—they No. 20-2243 29 counseled patients and prescribed drugs without supervision.” 136 S. Ct. at 1997 . The complaint thus alleged a falsehood by omission. The Court held that allegations of fraudulent omissions might suffice to state an FCA claim based on a theory of implied false certification. Id. at 1999 . In so holding, the Court described the paradigm case of implied false certification as follows: “When, as here, a defendant makes representations in submitting a claim but omits its violations of statutory, regulatory, or contractual requirements, those omissions can be a basis for liability if they render the defendant’s representations misleading with respect to the goods or services provided.” Id. Prose’s allegations are best conceptualized as a possible claim under a theory of implied falsehood. Following Escobar’s lead, we should not stretch the facial-falsehood concept but instead analyze the allegations under the rubric of implied false certification. C. Implied False Certification Turning now to the theory that is the closest fit with Prose’s allegations, I note for starters that my colleagues skip the threshold requirements announced in Escobar and instead move straight to the second-tier question of materiality. That is, the majority simply states, without explanation, that material omissions are implied false certifications. Majority op. at 14 (“Material omissions can suffice.”). That sweeping generalization cannot be squared with Escobar’s requirements for this type of FCA claim. Escobar held that a claim for payment might be an actionable violation of the FCA under a theory of an implied false 30 No. 20-2243 certification if two conditions are present: “first, the claim does not merely request payment[] but also makes specific representations about the goods or services provided; and second, the defendant’s failure to disclose noncompliance with material statutory, regulatory, or contractual requirements makes those representations misleading half-truths.” 136 S. Ct. at 2001 . Prose’s allegations do not satisfy the first of these threshold conditions. Molina’s enrollment forms did not make any specific representation about the goods or services provided. They simply enrolled Medicaid beneficiaries by rate cell, which designated the appropriate capitation rate for a given enrollee. This rate-cell information was nothing more than a request for a specific amount of payment for a very broad swath of services. In Escobar, by contrast, the medical contractor “submit[ed] claims for payment using payment codes that corresponded to specific counseling services.” 136 S. Ct. at 2000 (emphasis added). In other words, the claims for payment at issue in Escobar were specific claims that misled the government into believing something false. Here, Molina’s enrollment forms made broad payment requests covering a host of services, only one of which was not delivered. That does not satisfy Escobar’s first condition for a cognizable claim of implied false certification. Indeed, our own precedent confirms that a general request for payment coupled with some degree of contractual or regulatory noncompliance is not enough to support a claim of implied false certification. In United States v. SanfordBrown, Ltd. (“Sanford-Brown I”), 788 F.3d 696 (7th Cir. 2015), vacated United States ex rel. Nelson v. Sanford-Brown, Ltd., 136 S. Ct. 2506 (2016), reinstated in part by Sanford-Brown II, No. 20-2243 31 840 F.3d at 447 , we considered an FCA action brought against a for-profit college. The school signed a Program Participation Agreement with the Department of Education in which the college agreed to comply with all regulations under Title IV in exchange for federal subsidies. Id. at 707– 08. The college did not comply with all regulations, yet it submitted requests for funds anyway. Id. at 708. On remand from the Supreme Court, we held that the plaintiff had not satisfied Escobar’s first condition because he “offered no evidence that defendant Sanford-Brown College … made any representations at all in connection with its claims for payment, much less false or misleading representations.” Sanford-Brown II, 840 F.3d at 447 . In other words, a generic payment request—without specific representations about the goods or services provided—does not satisfy Escobar’s first condition and thus cannot suffice as an implied false certification. Escobar’s second condition requires the plaintiff to adequately allege (and later prove) that the defendant’s failure to disclose its noncompliance with a statutory or regulatory requirement made the specific representation a misleading half-truth. A half-truth is a “representation[] that state[s] the truth only so far as it goes, while omitting critical qualifying information.” Escobar, 136 S. Ct. at 2000 . Imagine that the Green Bay Packers have a bye week and someone makes the statement, “the Packers didn’t win today.” That’s a classic half-truth. The statement is true as far as it goes, but it directly implies a specific falsehood to an unaware fan: that the Packers lost that day. Escobar identified some helpful examples of half-truths. “A classic example of an actionable half-truth in contract law 32 No. 20-2243 is the seller who reveals that there may be two new roads near a property he is selling[] but fails to disclose that a third potential road might bisect the property.” Id. “Likewise, an applicant for an adjunct position at a local college makes an actionable misrepresentation when his resume lists prior jobs and then retirement[] but fails to disclose that his ‘retirement’ was a prison stint for perpetrating a $12 million bank fraud.” Id. Or consider the half-truth at issue in Escobar itself: the medical contractor’s submission of claims with payment codes and identification numbers corresponding to specific job titles and counseling services while not disclosing that the counselors providing the services were unlicensed. What we can distill from these examples is that a misleading half-truth arises when a defendant makes a specific statement (the Packers didn’t win today) that inevitably leads the recipient to assume by implication a particular falsehood (that the Packers lost). Prose’s allegations operate at a much higher level of generality than the allegations in Escobar. In that case there was a tight link between the specific representations (payment codes for counseling services and ID numbers for job titles) and the falsehood inevitably implied by omission (the counselors corresponding to the identified job titles were in fact licensed for those positions). Here, there is at most only a loose association between Molina’s nonspecific representation (enrolling a Medicaid beneficiary in the nursing-facility rate cell) and the alleged false implication (that SNFist services—one among many nursing-facility services—were actually provided). Where, as here, the defendant’s claim for payment wouldn’t necessarily lead the recipient to assume the specific falsehood alleged in the complaint, there is no half-true statement and thus no falsehood by implication. No. 20-2243 33 Indeed, Molina’s enrollment forms made no specific representations about the services provided beyond enrolling a beneficiary in a given rate cell, which after all, is just a request for a certain payment amount. Considering the multitude of services provided to nursing-home enrollees, the enrollment forms wouldn’t inevitably lead the government to assume any specific falsehood by implication. The enrollment forms, though perhaps misleading in a general sense, did not contain a specific half-true statement as required by Escobar. * * * The majority concentrates its implied-falsehood analysis on the question of materiality, an additional requirement for a viable FCA claim and one that Escobar also addressed at some length. A representation is material if “a reasonable man would attach importance to [it] in determining his choice of action in the transaction” or if “the defendant knew or had reason to know that the recipient of the representation attaches importance to the specific matter.” Id. at 2002– 03 (quotation marks omitted). My colleagues rely almost entirely on the difference in capitation rates among rate cells: $3,127 per month for a nursing-facility enrollee; about $2,500 per month for a middle-tier enrollee; and $54 for a low-tier enrollee. They conclude that “[t]he size of the price differential alone offers strong support for a finding of materiality.” Majority op. at 15; see also id. at 17 (“The contract itself, which fixes the cost of the NF category well above the other tiers, is powerful evidence of the materiality of the SNF services.”). But by omitting SNFist services, Molina didn’t provide middle-tier service at high-tier rates. Instead, it provided something 34 No. 20-2243 close to high-tier service at high-tier rates. By itself, the difference in capitation rates sheds little light on the materiality question because nothing in the complaint connects that difference to SNFist services. In some cases a large pay differential between two billing rates might alone be enough to support an inference of materiality. Not so here. The problem turns again on the nature of SNFist services. To repeat, SNFist services are carecoordination services—one of many services provided to nursing-home enrollees that in the aggregate contribute to the higher capitation rate. The complaint offers nothing to explain the effect of these particular services on the government’s willingness to pay the nursing-facility capitation rate for these enrollees. Without some factual contextualization, we cannot draw an inference that Molina’s nondisclosure was material to the government’s decision to pay its claims during the relevant time period. Think of it this way: If rate cell 1 corresponds to 10 services provided at a rate of $2,000 and rate cell 2 corresponds to those same 10 services plus SNFist services at a rate of $3,000, then billing at the level 2 rate while not providing SNFist services would support an inference of materiality at the pleading stage. If SNFist services are not delivered, then the contractor is providing only level 1 services, and a reasonable person would not pay much higher level 2 rates for receiving only level 1 services. But now consider a scenario in which rate cell 2 corresponds to 30 services—the 10 in rate cell 1 plus 20 others, one of which is SNFist services. In that scenario it doesn’t make sense to rely on the $1,000 price differential in considering whether the omission of SNFist services is material No. 20-2243 35 because the differential may be largely explained by the 19 other services separating rate cell 1 and rate cell 2. That’s the situation here—the difference in capitation rates between the nursing-home rate and the middle-tier rate is only partially explained by SNFist services, and nothing in the complaint illuminates the extent to which those services account for the differential. Without at least some contextualizing factual allegations, the capitation-rate differentials are not a useful metric for assessing materiality. Of course, materiality might be established in other ways, but Prose’s remaining arguments are unpersuasive; even the majority doesn’t make use of them. For example, he points to the fact that the government specifically discussed SNFist services during 2013 contract negotiations and asks us to infer that they were material to the government’s decision to pay Molina in 2015, 2016, and 2017. But the mere discussion of a contract term earlier in negotiations doesn’t mean that its fulfillment is material to a later decision to pay, especially when the negotiations occurred years before. Prose also asks us to infer materiality because SNFist services were supposed to be available 24/7 and were coupled with reporting obligations. But the contract requires every covered service to be provided 24/7 and is replete with reporting obligations, which undermines any suggestion that SNFist services had special status. Finally, Prose argues that SNFist services are necessarily material because payment rates are derived from actuarially precise calculations that included them. This reasoning suggests that every service under a contract with actuarial pricing is material. That’s an unsound approach to the materiality question in this context. Although the contract 36 No. 20-2243 may have calibrated the capitation rates to the services the government expected to be delivered, it doesn’t follow that the government would withhold payment if a single one of those services wasn’t provided. Escobar characterized the materiality standard as “demanding,” 136 S. Ct. at 2003, and Prose has failed to meet it. Perhaps he could have done so with factual allegations showing that SNFist services account to a significant degree for the difference in capitation rates. Or perhaps he could have alleged that Molina was aware that the government “consistently refuses to pay claims in the mine run of cases” if SNFist services are omitted. Id. But we know the opposite is true, as my colleagues acknowledge. Majority op. at 16–17 (explaining that “the government not only continued paying [Molina] after Prose brought this case, but it also renewed its contract with Molina twice during that time”). Escobar explained that “if the [g]overnment regularly pays a particular type of claim in full despite actual knowledge that certain requirements were violated, and has signaled no change in position, that is strong evidence that the requirements are not material.” Id. at 2003–04. As it is, we’re left with only generic statements about the importance of SNFist services and a rate differential without any contextualizing factual allegations connecting the differential to the omitted services. That doesn’t clear the bar. Even if my analysis of materiality is wrong, the majority’s conclusion that Prose has stated a claim for implied false certification essentially establishes a new rule that any claim for payment while in material noncompliance with a contract or governing law is an actionable violation of the FCA. No. 20-2243 37 As already explained, that conclusion conflicts with Escobar and circuit caselaw. * * * For these reasons, I would affirm the judgment dismissing the complaint for failure to state a claim. Prose’s allegations fall short of satisfying Rule 9(b)’s heightened pleading standard for an actionable FCA claim under any of the three recognized theories of liability.