UNITED STATES of America, EX REL. J. William BOOKWALTER, III, M.D.; Robert J. Sclabassi, M.D.; Anna Mitina v. UPMC; University of Pittsburgh Physicians, d/b/a UPP Department of Neurosurgery J. William Bookwalter, III, M.D.; Robert J. Sclabassi, M.D.; Anna Mitina, Appellants
OPINION OF THE COURT
Table of Contents
A. Factual Background․166
1. The University of Pittsburgh medical system․166
2. The neurosurgeons' compensation structure․166
3. The neurosurgeons' alleged fraud and its effects on salaries and revenues․167
B. Procedural History․167
II. Standards of Review and Pleading․168
III. The Stark Act and the False Claims Act․168
A. The Stark Act․168
1. Forbidden conduct․168
3. No built-in cause of action․169
B. The False Claims Act․169
IV. The Relators Plead Stark Act Violations․169
A. The surgeons referred designated health services to the hospitals․170
B. The relators' complaint alleges an indirect compensation arrangement․170
1. An unbroken chain of entities with financial relationships connects the surgeons with the hospitals․171
2. The surgeons' compensation took into account the volume and value of their referrals․171
3. The hospitals knew that the surgeons' compensation took their referrals into account․174
V. The Relators Plead False Claims Act Violations․175
A. The pleadings satisfy all three elements of the False Claims Act․175
B. The pleadings satisfy Rule 9(b)․176
C. Pleading Stark Act exceptions under the False Claims Act․177
1. The burden of pleading Stark Act exceptions stays with the defendant under the False Claims Act․177
2. Even if the relators bore this pleading burden, they have met it․177
Healthcare spending is a huge chunk of the federal budget. Medicare and Medicaid cost roughly a trillion dollars per year. And with trillions of dollars comes the temptation for fraud.
Fraud is a particular danger because doctors and hospitals can make lots of money for one another. When doctors refer patients to hospitals for services, the hospitals make money. There is nothing inherently wrong with that. But when hospitals pay their doctors based on the number or value of their referrals, the doctors have incentives to refer more. The potential for abuse is obvious and requires scrutiny.
The Stark Act and the False Claims Act work together to ensure this scrutiny and safeguard taxpayer funds against abuse. The Stark Act forbids hospitals to bill Medicare for certain services when the hospital has a financial relationship with the doctor who asked for those services, unless an exception applies. And the False Claims Act gives the government and relators a cause of action with which to sue those who violate the Stark Act.
Here, the relators allege that the defendants have for years been billing Medicare for services referred by their neurosurgeons in violation of the Stark Act. The District Court found that the relators had failed to state a plausible claim and dismissed their suit.
This appeal revolves around two questions: First, do the relators offer enough facts to plausibly allege that the surgeons' pay varies with, or takes into account, their referrals? Second, who bears the burden of pleading Stark Act exceptions under the False Claims Act?
The answer to the first question is yes. The relators' complaint alleges enough facts to make out their claim. The relators make a plausible case that the surgeons' pay is so high that it must take their referrals into account. All these facts are smoke; and where there is smoke, there might be fire.
The answer to the second question is the defendants. The Stark Act's exceptions work like affirmative defenses in litigation. The burden of pleading these affirmative defenses lies with the defendant. This is true even under the False Claims Act. And even if that burden lay with the relators, their pleadings meet that burden here.
We hold that the complaint states plausible violations of both the Stark Act and the False Claims Act. So we will reverse.
A. Factual Background
1. The University of Pittsburgh medical system. On this motion to dismiss, we take as true the facts alleged in the second amended complaint: The University of Pittsburgh Medical Center is a multi-billion-dollar nonprofit healthcare enterprise. The Medical Center is the parent organization of a whole system of healthcare subsidiaries, including twenty hospitals. The Medical Center is the sole member (owner) of each hospital.
More than 2,700 doctors, including dozens of neurosurgeons, work at these hospitals. The doctors are employed not by the hospitals, but by other Medical Center subsidiaries. Three of these subsidiaries matter here: University of Pittsburgh Physicians; UPMC Community Medicine, Inc.; and Tri-State Neurological Associates-UPMC, Inc.
These three subsidiaries employed many of the neurosurgeons who worked at the Medical Center's hospitals during the years at issue, from 2006 on. Pittsburgh Physicians' Neurosurgery Department employed most of the surgeons at issue. Tri-State employed two, and Community Medicine employed one. The Medical Center owns all three subsidiaries. In short, the Medical Center owns both the hospitals and the companies that employ the surgeons who work in the hospitals.
2. The neurosurgeons' compensation structure. The surgeons who worked for the three subsidiaries here all had similar employment contracts. Each surgeon had a base salary and an annual Work-Unit quota. Work Units (or wRVUs) measure the value of a doctor's personal services. Every medical service is worth a certain number of Work Units. The longer and more complex the service, the more Work Units it is worth. Work Units are one component of Relative Value Units (RVUs). RVUs are the basic units that Medicare uses to measure how much a medical procedure is worth.
The surgeons were rewarded or punished based on how many Work Units they generated. If a surgeon failed to meet his yearly quota, his employer could lower his future base salary. But if he exceeded his quota, he earned a $45 bonus for every extra Work Unit.
3. The neurosurgeons' alleged fraud and its effects on salaries and revenues. This compensation structure gave the surgeons an incentive to maximize their Work Units. And the incentive seems to have worked. The surgeons reported doing more, and more complex, procedures. So the number of Work Units billed by the Neurosurgery Department more than doubled between 2006 and 2009.
Much of this increase allegedly stemmed from fraud. The relators accuse the surgeons of artificially boosting their Work Units: The surgeons said they acted as assistants on surgeries when they did not. They said they acted as teaching physicians when they did not. They billed for parts of surgeries that never happened. They did surgeries that were medically unnecessary or needlessly complex. And they did these things, say the relators, “[w]ith the full knowledge and endorsement of” the Medical Center. App. 184 ¶190.
Fraud can be profitable. And here it allegedly was. With these practices, the surgeons racked up lots of Work Units and made lots of money. Most reported total Work Units that put them in the top 10% of neurosurgeons nationwide. And some received total pay that put them among the best-paid 10% of neurosurgeons in the country.
The surgeons' efforts proved profitable for the Medical Center too. The Medical Center made money off the surgeons' work on some of the referrals. And to boot, healthcare providers bill Medicare for more than just the surgeons' own Work Units. Whenever a surgeon did a procedure at one of the hospitals, the Medical Center also got to bill “for the attendant hospital and ancillary services.” App. 166 ¶ 104. This part of the bill could be four to ten times larger than the cost of the surgeon's own services. So when the surgeons billed more, the Medical Center made more. “Indeed, in 2009,” the Neurosurgery Department “was the single highest grossing neurosurgical department in the United States, with Medicare charges alone of $58.6 million.” App. 163–64 ¶91.
B. Procedural History
The relators first filed suit in 2012. They alleged that the Medical Center, Pittsburgh Physicians, and a bevy of neurosurgeons had submitted false claims for physician services and for hospital services to Medicare and Medicaid. Four years later, the United States intervened as to the claims for physician services. The government settled those claims for about $2.5 million. It declined to intervene as to the claims for hospital services, but it let the relators maintain that part of the action in its stead.
After the government intervened, the District Court dismissed the first amended complaint without prejudice for failure to state a claim. The relators then filed their current complaint, asserting three causes of action against the Medical Center and Pittsburgh Physicians under the False Claims Act:
(1) one count of submitting false claims,
(2) one count of knowingly making false records or statements, and
(3) one count of knowingly making false records or statements material to an obligation to pay money to the United States.
The District Court again dismissed for failure to state a claim, this time with prejudice. The relators now appeal.
II. Standards of Review and Pleading
We review a district court's dismissal for failure to state a claim de novo. Vorchheimer v. Philadelphian Owners Ass'n, 903 F.3d 100, 105 (3d Cir. 2018). Our job is to gauge whether the complaint states a plausible claim to relief. Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Plausible does not mean possible. If the allegations are “merely consistent with” misconduct, then they state no claim. Bell Atl. Corp. v. Twombly, 550 U.S. 544, 557, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007). There must be something in the complaint to suggest that the defendant's alleged conduct is illegal. Id. at 557, 127 S.Ct. 1955.
But plausible does not mean probable either. Our job is not to dismiss claims that we think will fail in the end. See id. at 556, 127 S.Ct. 1955. Instead, we ask only if we have “enough fact[s] to raise a reasonable expectation that discovery will reveal evidence of” each element. Id.
This is the baseline pleading standard for all civil actions. Fed. R. Civ. P. 8; Iqbal, 556 U.S. at 684, 129 S.Ct. 1937. But the relators allege claims for fraud. So they must also meet Rule 9(b)'s heightened pleading requirement. United States ex rel. Moore & Co., P.A. v. Majestic Blue Fisheries, LLC, 812 F.3d 294, 306–07 (3d Cir. 2016). That rule says that a party alleging fraud “must state with particularity the circumstances constituting fraud.” Fed. R. Civ. P. 9(b).
III. The Stark Act and the False Claims Act
A. The Stark Act
The Stark Act and its regulations broadly bar Medicare claims for many services referred by doctors who have a financial interest in the healthcare provider. But the statute creates dozens of exceptions and authorizes the Department of Health and Human Services to make even more exceptions for financial relationships that “do[ ] not pose a risk of program or patient abuse.” 42 U.S.C. § 1395nn(b)(4).
1. Forbidden conduct. The Stark Act opens with a broad ban. It forbids submitting Medicare claims for “designated health services” provided under a “referral” made by a doctor with whom the entity has a “financial relationship.” Id. § 1395nn(a)(1). Understanding this ban requires exploring these three quoted terms, each of which has statutory and regulatory definitions.
The Stark Act lists several categories of designated health services, including inpatient hospital services. Id. § 1395nn(h)(6)(K). And inpatient hospital services include bed and board, interns' and residents' services, nursing, drugs, supplies, transportation, and overhead. 42 C.F.R. §§ 409.10(a), 411.351.
A referral is a doctor's request for a designated health service. 42 U.S.C. § 1395nn(h)(5)(A); 42 C.F.R. § 411.351. That definition is broad, but it has an important exception: services that a doctor performs personally do not count. 42 C.F.R. § 411.351. That makes sense; ordinarily, one cannot refer something to oneself. And the exception's boundaries also follow: it does not cover services by a doctor's associates or employees, or services incidental to the doctor's own services. Id.; Medicare Program; Physicians' Referrals to Health Care Entities with Which They Have Financial Relationships (Phase II); Interim Final Rule, 69 Fed. Reg. 16054, 16063 (Mar. 26, 2004).
Finally, financial relationships come in two forms: (1) ownership or investment interests and (2) compensation arrangements. 42 U.S.C. § 1395nn(a)(2). This case turns on the latter. The statute defines compensation arrangement to mean “any arrangement involving any remuneration between” a doctor and a healthcare provider. Id. § 1395nn(h)(1)(A). And remuneration “includes any remuneration, directly or indirectly, in cash or in kind.” Id. § 1395nn(h)(1)(B).
2. Exceptions. On its face, the Stark Act's ban sweeps in lots of common situations. To separate the wheat from the innocuous chaff, Congress and the Department of Health and Human Services have created many exceptions. Here, the Medical Center argues that exceptions for four types of compensation arrangements could apply here: bona fide employment; personal services; fair-market-value compensation; and indirect compensation. See id. § 1395nn(e)(2), (e)(3); 42 C.F.R. § 411.357(l), (p).
All four exceptions have two elements in common. First, the doctor's compensation must not “take[ ] into account (directly or indirectly) the volume or value of” the doctor's referrals. 42 U.S.C. § 1395nn(e)(2)(B)(ii); accord id. § 1395nn(e)(3)(A)(v); 42 C.F.R. § 411.357(l)(3), (p)(1)(i). Second, the doctor's compensation must not exceed fair market value. 42 U.S.C. § 1395nn(e)(2)(B)(i), (e)(3)(A)(v); 42 C.F.R. § 411.357(l)(3), (p)(1)(i).
In litigation, these exceptions are affirmative defenses. So once a plaintiff proves a prima facie violation of the Stark Act, the burden shifts to the defendant to prove that an exception applies. United States ex rel. Kosenske v. Carlisle HMA, Inc., 554 F.3d 88, 95 (3d Cir. 2009).
3. No built-in cause of action. The Stark Act forbids the government to pay claims that violate the Act. 42 U.S.C. § 1395nn(g)(1). It demands restitution from those who receive payments on illegal claims. Id. § 1395nn(g)(2). And it creates civil penalties for submitting improper claims or taking part in schemes to violate the Act. Id. § 1395nn(g)(3), (4). But it gives no one a right to sue. United States ex rel. Drakeford v. Tuomey, 792 F.3d 364, 374 n.4 (4th Cir. 2015).
So the Stark Act never appears in court alone. Instead, it always come in through another statute that creates a cause of action—typically, the False Claims Act.
B. The False Claims Act
Under the False Claims Act, any person who “knowingly presents, or causes to be presented, a false or fraudulent claim for payment or approval” is civilly liable to the United States. 31 U.S.C. § 3729(a)(1)(A). A Medicare claim that violates the Stark Act is a false claim under the False Claims Act. Kosenske, 554 F.3d at 94. The False Claims Act also makes liable anyone who “knowingly makes, uses, or causes to be made or used, a false record or statement material to” a false or fraudulent claim. 31 U.S.C. § 3729(a)(1)(B), (G).
IV. The Relators Plead Stark Act Violations
A prima facie Stark Act violation has three elements: (1) a referral for designated health services, (2) a compensation arrangement (or an ownership or investment interest), and (3) a Medicare claim for the referred services. See United States ex rel. Schmidt v. Zimmer, Inc., 386 F.3d 235, 241 (3d Cir. 2004). This combination of factors suggests potential abuse of Medicare. When they are all present, we let plaintiffs go to discovery.
Here, no one denies that the defendants made Medicare claims for designated health services. The issue is whether the complaint sufficiently alleges referrals and a compensation arrangement. We hold that it does. The alleged Medicare abuse is plausible and deserves more scrutiny.
A. The surgeons referred designated health services to the hospitals
The relators allege that “[e]very time [the neurosurgeons] performed a surgery or other procedure at the UPMC Hospitals, [they] made a referral for the associated hospital claims.” App. 193 ¶234. They are right that these claims are referrals.
As mentioned, the law defines referrals broadly. A referral is a doctor's request for any designated health service that is covered by Medicare and provided by someone else. 42 C.F.R. § 411.351. Designated health services include bed and board, some hospital overhead, nursing services, and much more. 42 C.F.R. § 409.10(a). And the relators plead that as the surgeons performed more procedures, those procedures required (and the hospital provided and “increased billings for[) ] the attendant hospital and ancillary services including ․ hospital and nursing charges.” App. 166 ¶104 (emphasis added). So the plaintiffs plead that the surgeons referred designated health services to the hospitals.
Treating these services as referrals makes sense. The Stark Act's first step is to flag all potentially abusive arrangements. And doctors who generate profits for a hospital may be tempted to abuse their power, raising hospital bills as well as their own pay. These financial arrangements thus deserve a closer look. And they will get a closer look only if we call these arrangements what they are: doctors referring services to hospitals.
The Department of Health and Human Services agrees. In Phase I of its Stark Act rulemaking, it considered this point. It determined that “any hospital service, technical component, or facility fee billed by [a] hospital in connection with [a doctor's] personally performed service” counts as a referral. Medicare and Medicaid Programs; Physicians' Referrals to Health Care Entities with Which They Have Financial Relationships, 66 Fed. Reg. 856, 941 (Jan. 4, 2001). This is true even “in the case of an inpatient surgery” where the doctor performs the surgery. Id.
Then, in Phase II of its rulemaking, the agency revisited the question and considered narrower definitions. For instance, many commenters suggested excluding “services that are performed ‘incident to’ a physician's personally performed services or that are performed by a physician's employee” from the definition of a referral. 69 Fed. Reg. at 16063.
But the agency reasonably rejected these suggestions. A narrower view, it reasoned, would all but swallow at least one statutory exception. Id. And it explained that the availability of that and other exceptions did enough to protect innocent conduct. Id. “[T]his interpretation is consistent with the statute as a whole,” which begins by casting a broad net to scrutinize all potential abuse. Id.
B. The relators' complaint alleges an indirect compensation arrangement
A referral is ripe for abuse only when the doctor who made it has a financial relationship with the provider. Only then can a doctor profit from his own referral. The financial relationship here is a compensation arrangement.
Compensation arrangements can be either direct or indirect. 42 C.F.R. § 411.354(c). The hospitals did not pay the surgeons directly. So if there is any compensation arrangement here, it is indirect. That requires three elements: First, there must be “an unbroken chain ․ of persons or entities that have financial relationships” connecting the referring doctor with the provider of the referred services. Id. § 411.354(c)(2)(i). Second, the referring doctor must get “aggregate compensation ․ that varies with, or takes into account, the volume or value of referrals.” Id. § 411.354(c)(2)(ii). And third, the service provider must know, recklessly disregard, or deliberately ignore that the doctor's compensation “varies with, or takes into account, the volume or value of referrals.” Id. § 411.354(c)(2)(iii). (The parties do not challenge any of the regulations at issue, so we likewise assume that they are valid.) The complaint plausibly pleads enough facts to satisfy each element.
1. An unbroken chain of entities with financial relationships connects the surgeons with the hospitals. An unbroken chain of financial relationships links the surgeons to the hospitals. First, the Medical Center owns each hospital. Second, the Medical Center also owns three entities: Pittsburgh Physicians, Community Medicine, and Tri-State. Third, each of these three entities employs and pays at least one of the surgeons. That adds up to an unbroken chain of financial relationships. Neither party disputes this.
2. The surgeons' suspiciously high compensation suggests that it took into account the volume and value of their referrals. Next, the relators allege that the surgeons' aggregate compensation varied with, and took into account, their referrals.
The parties disagree about what it means for compensation to vary with referrals. Appellants argue that varies with requires only correlation. And compensation correlates with referrals here, they argue, because surgeons racked up more Work Units and earned more money by generating more referrals. So the surgeons' aggregate compensation allegedly varied with their referrals. Appellees, by contrast, deny that a correlation suffices. Rather, they insist that the law requires some form of causation.
We need not resolve the meaning of varies with here. Regardless, the complaint plausibly alleges that the surgeons' compensation takes into account the volume or value of their referrals. Under the Stark Act and its regulations, compensation takes into account referrals if there is a causal relationship between the two. And here, the surgeons' suspiciously high compensation suggests causation.
Compensation for personal services above the fair market value of those services can suggest that the compensation is really for referrals. This is just common sense. Healthcare providers would not want to lose money by paying doctors more than they bring in. They would do so only if they expected to make up the difference another way. And that way could be through the doctors' referrals.
This may not be obvious on the face of the statute and regulations. The Stark Act often treats fair market value as a concept distinct from taking into account the volume or value of referrals. For example, these two concepts are separate elements of many Stark Act exceptions. E.g., 42 U.S.C. § 1395nn(e)(2) (bona fide employment), (e)(3) (personal service); 42 C.F.R. § 411.357(l) (fair-market-value compensation), (p) (indirect compensation). And the definition of an indirect compensation arrangement includes taking referrals into account, but not fair market value. 42 C.F.R. § 411.354(c)(2)(ii).
But the Act's different treatment of these concepts does not sever them. To start, just because a statute has two elements does not mean that one can never be evidence of the other. Theft requires taking another's property with intent. Those are two elements, but the fact of taking property can be circumstantial evidence of intent.
So too here. Perhaps not all payments above fair market value are evidence of taking into account the doctor's referrals. But common sense says that marked overpayments are a red flag. Anyone would wonder why the hospital would pay so much if it was not taking into account the doctor's referrals for other services. And we do no violence to the statutory text by seeking an answer to that question.
The agency confronted this question directly. It remarked that even “fixed aggregate compensation can form the basis for a prohibited ․ indirect compensation arrangement” if it “is inflated to reflect the volume or value of a physician's referrals.” 69 Fed. Reg. at 16059 (emphasis added). The same is true of “unit-of-service-based compensation arrangements,” like the one here. Id. Excessive compensation is thus a sign that a surgeon's pay in fact takes referrals into account.
So aggregate compensation that far exceeds fair market value is smoke. It suggests that the compensation takes referrals into account. And the relators here plead five facts that, viewed together, make plausible claims that the surgeons' pay far exceeded their fair market value. First, some surgeons' pay exceeded their collections. Second, many surgeons' pay exceeded the 90th percentile of neurosurgeons nationwide. Third, many generated Work Units far above industry norms. Fourth, the surgeons' bonus per Work Unit exceeded what the defendants collected on most of those Work Units. And finally, the government alleged in its settlement agreement that the Medical Center had fraudulently inflated the surgeons' Work Units. That much smoke makes fire plausible.
a. Pay exceeding collections. Paying a worker more than he brings in is suspicious. And the complaint alleges that at least three surgeons (Drs. Bejjani, Spiro, and El-Kadi) were paid more than the Medical Center collected for their services. The complaint also alleges that the Medical Center credits surgeons with 100 percent of the Work Units that they generate, even if it cannot collect on all of them. So at least three surgeons (maybe more) were paid more than they bring in.
b. Pay exceeding the 90th percentile. The relators allege that “[c]ompensation exceeding the 90th percentile is widely viewed in the industry as a ‘red flag’ indicating that it is in excess of fair market value.” App. 191 ¶223. The defendants do not deny this.
Several surgeons were paid more than the 90th percentile. For example, the relators point to the compensation of Drs. Abla, Spiro, Kassam, and Bejjani between 2008 and 2011. Apart from Dr. Spiro in 2008, each of these surgeons was paid more than even the highest estimate of the 90th percentile for all U.S. neurosurgeons in all four years. And depending on which estimate of the 90th percentile you use, they were sometimes paid two or three times more than the 90th percentile. Dr. Bejjani's 2011 bonus alone exceeded the 90th percentile of total compensation in some surveys.
c. Extreme Work Units. The relators also allege facts from which we can reasonably infer that the surgeons generated far more Work Units than normal. Many neurosurgeons “were routinely generating [Work Units] exceeding by an enormous margin the 90th percentile as reflected in widely-accepted market surveys.” App. 171 ¶126. Even if we look only at the highest industry estimates, all but one of the surgeons reported Work Units above the 90th percentile in 2006 and 2007. In 2008 and 2009, eight of the twelve named surgeons exceeded the highest estimate of the 90th percentile. A few even seemed “super human,” racking up two to three times the 90th percentile. App. 169 ¶ 117.
In short, most of the surgeons generated Work Units at or above the 90th percentile. Some of their numbers were unbelievably high. And because their pay depends in large part on their Work Units, it is fair to infer that most of their pay was also at or above the 90th percentile.
d. Bonuses exceeding the Medicare reimbursement rate. Once a surgeon had enough Work Units to earn bonus pay, the bonus per Work Unit was more than Medicare would pay for each one. The surgeons' bonus per Work Unit was $45. But the Medicare reimbursement rate was only about $35. So once surgeons became eligible for bonuses, the defendants took an immediate loss on every Work Unit submitted to Medicare.
On its own, this would not show that the surgeons were overpaid. Medicare and Medicaid are well known as bottom-billers. They pay less than private insurers. Though the defendants lost some money on Medicare Work Units, perhaps they made it back with Work Units billed to other insurers.
But the relators also allege that “the majority of all claims submitted by the [defendants] ․ were submitted to federal health insurance programs such as Medicare and Medicaid.” App. 193 ¶233. We cannot assume that private payments suffice to make up the difference. Doing so would disregard our job at this stage: to draw reasonable inferences in favor of the plaintiffs.
In short, the defendants took an immediate financial hit on Work Units for a majority of their claims. This is yet another sign that the surgeons' pay took referrals into account.
The defendants disagree. They argue that the surgeons earn high salaries because of bona fide bargaining with their employers. Their salaries supposedly represent the market's demand for their surgical skill and experience.
This argument fails for two reasons. First, the complaint says nothing about the surgeons' skill and experience or the Pittsburgh market for surgeons. On this motion to dismiss, we cannot go beyond the well-pleaded facts in the complaint.
Second, a bare claim of bona fide bargaining is not enough. The Stark Act recognizes that related parties often negotiate agreements “to disguise the payment of non-fair-market-value compensation.” Kosenske, 554 F.3d at 97. We trust that bona fide bargaining leads to fair market value only when neither party is “in a position to generate business for the other.” Id.; 42 C.F.R. § 411.351 (defining “fair market value” and “general market value”). But that is not true here. The surgeons and the Medical Center can generate business for each other. So we cannot assume that any bargaining was bona fide or that the resulting pay was at fair market value.
e. The possibility of fraud. Finally, the surgeons' high pay may have been based on fudging the numbers. Not only were their individual Work Units “significantly out of line with industry benchmarks,” but the Neurosurgery Department as a whole realized astounding “annual growth rates of work [Units] ․ of 20.3%, 57.1% and 20.0%” in 2007, 2008, and 2009. App. 171 ¶¶127–28. Two of the surgeons more than doubled their output in just a few years. The relators allege that the defendants got this growth by “artificially inflat[ing] the number of [Work Units] in a number of ways.” App. 171 ¶130.
Alleging this fraud, the relators' first complaint included claims “relating to physician services submitted by” the defendants along with the “hospital claims” currently before us. App. 189 ¶217 (emphases in original) The government chose to intervene as to the former claims, settling them with the defendants for almost $2.5 million.
The relators' current complaint quotes that settlement agreement. In it, the government accused the surgeons of many fraudulent practices: They claimed to have acted as assistants when they did not. They claimed to have done more extensive surgeries than they did. And they chose the wrong codes for surgeries. So “claims submitted for these physician services resulted in more reimbursement than would have been paid” otherwise. App. 188–89 ¶216.
We are careful not to overstate the point. This settlement is not an admission of guilt. It proves no wrongdoing. But at the 12(b)(6) stage, we are looking only for plausible claims, not proof of wrongs. And the government's choice to intervene after years of investigation and its allegations in the settlement are cause for suspicion.
The question is not whether a doctor was able to use an otherwise-valid compensation scheme as a vehicle for fraudulent billing. Not every fraudulent Medicare bill made at a hospital will give rise to a Stark Act violation. Here, however, where the compensation scheme produced results bordering on the absurd, relators plausibly assert that the system may have been designed with that outcome in mind.
The relators allege five sets of facts that suggest that the surgeons' pay far exceeded fair market value: pay exceeding collections, pay above the 90th percentile, extreme Work Units, bonuses above the Medicare reimbursement rate, and the settlement. That is plenty of smoke. We need not decide whether any of these allegations alone would satisfy the relators' pleading burden. Together, they plausibly suggest that the surgeons' pay took their referrals into account. Thus, the relators have pleaded more than enough facts to suggest an indirect compensation arrangement.
3. The hospitals knew that the surgeons' compensation took their referrals into account. The final element of an indirect compensation arrangement is scienter. To show scienter, the relators' pleadings must allege that the hospitals that provided the referred services either (1) knew, (2) deliberately ignored, or (3) recklessly disregarded that the surgeons got “aggregate compensation that varie[d] with, or t[ook] into account, the volume or value of referrals.” 42 C.F.R. § 411.354(c)(2)(iii). They allege this too.
To begin, the Medical Center controls all the hospitals and the surgeons' direct employers. It owns each hospital. And it owns Pittsburgh Physicians, Community Medicine, and Tri-State. So the Medical Center “has unfettered authority with respect to most members of the [medical system] and significant authority (including with respect to financial and tax matters) with respect to the remaining members.” App. 146–47 ¶19 (quoting a Medical Center tax filing).
Further, many officers and board members of these entities overlapped. For example, one person simultaneously served as an executive vice president of the Medical Center as well as the president and a board member of Pittsburgh Physicians. And he signed surgeons' pay agreements for Pittsburgh Physicians. The relators identify nine others who served on the board of both the Medical Center and another entity in the medical system. Authority was so centralized that a single person signed a settlement agreement on behalf of all the defendants that were part of the medical system. And with common control comes common knowledge.
The common knowledge included both the surgeons' pay and their referrals. The Medical Center took part in forming, approving, and implementing the surgeons' pay packages. So it knew their structure. The Medical Center also had a central coding and billing department that handled billing for its subsidiaries. So it knew about the surgeons' referrals.
With both sets of data in front of it, we can plausibly infer that the Medical Center knew the surgeons' compensation took their referrals into account. And as the Medical Center knew that, so did the hospitals. They had all the data right in front of them. They knew that the surgeons' pay and Work Units were out of line with industry survey data. Even if they did not actually know that the surgeons' pay and work levels were suspiciously high, they at least deliberately ignored or recklessly disregarded that fact. Thus, the complaint alleges that both the Medical Center and hospitals had scienter.
* * * * *
This means that the relators have successfully pleaded the third and final element of a Stark Act violation: scienter. But they must plead one more thing to survive a motion to dismiss. We must now consider whether the relators have pleaded a plausible prima facie case under the False Claims Act.
V. The Relators Plead False Claims Act Violations
The relators plead their Stark Act claims as violations of the False Claims Act. So their pleadings must satisfy all the elements of the False Claims Act. They do. And they satisfy Rule 9(b)'s heightened pleading standard. Last, we hold that the Stark Act's exceptions are not additional elements of a prima facie case. But even if they were, the relators have plausibly pleaded that no exception applies here.
A. The pleadings satisfy all three elements of the False Claims Act
To make out a prima facie case, the relators must plead three elements: “ ‘(1) the defendant presented or caused to be presented to an agent of the United States a claim for payment; (2) the claim was false or fraudulent; and (3) the defendant knew the claim was false or fraudulent.’ ” Schmidt, 386 F.3d at 242 (quoting Hutchins v. Wilentz, Goldman & Spitzer, 253 F.3d 176, 182 (3d Cir. 2001)). They have alleged enough facts to plead all three elements.
First, by submitting claims to Medicare and other federal health programs, the defendants presented claims for payment to the government.
Second, the relators allege that these claims were false. A Medicare claim that violates the Stark Act is a false claim. Kosenske, 554 F.3d at 94. And we have already explained at length why the Medicare claims here plausibly violated the Stark Act.
Third, the relators' allegations plead scienter. Just like the Stark Act, the False Claims Act requires that the defendants know, deliberately ignore, or recklessly disregard the falsity of their claim. 31 U.S.C. § 3729(b)(1)(A). But it does not require a specific intent to defraud. Id. § 3729(b)(1)(B).
The claims are false because they allegedly violated the Stark Act. The question is whether the defendants at least recklessly disregarded that possibility. The defendants had a centralized billing department and were familiar with the Stark Act itself, so they knew that they submitted Medicare claims for referred designated health services. That leaves only whether the defendants knew that the hospitals and surgeons had an indirect compensation agreement.
The complaint alleges that the defendants at least recklessly disregarded that possibility. They knew their own corporate structure. We have already explained how they knew or recklessly disregarded that the surgeons' pay varied with their referrals. And we have also explained how they knew or recklessly disregarded that their surgeons' pay far exceeded fair market value and thus plausibly took referrals into account. So the relators have pleaded a prima facie claim under the False Claims Act.
B. The pleadings satisfy Rule 9(b)
The relators' complaint also satisfies Rule 9(b)'s particularity requirement. To do so, the allegations must go well beyond Rule 8's threshold of plausibility. A mere plausible inference of illegality is not enough. Instead, “a relator must ‘establish a “strong inference” that the false claims were submitted.’ ” United States ex rel. Silver v. Omnicare, Inc., 903 F.3d 78, 92 (3d Cir. 2018) (quoting Foglia v. Renal Ventures Mgmt., 754 F.3d 153, 158 (3d Cir. 2014)).
Rule 9(b)'s particularity requirement requires a plaintiff to allege “ ‘all of the essential factual background that would accompany the first paragraph of any newspaper story—that is, the who, what, when, where, and how of the events at issue.’ ” Majestic Blue Fisheries, 812 F.3d at 307 (quoting In re Rockefeller Ctr. Props., Inc. Secs. Litig., 311 F.3d 198, 217 (3d Cir. 2002)). The complaint gives us all these necessary details:
• Who? The defendants: the Medical Center and Pittsburgh Physicians.
• What? The defendants submitted or caused to be submitted false Medicare claims.
• When? From 2006 until now.
• Where? The Medicare claims were submitted from the Medical Center's centralized billing facility, while the referred services were provided at the Medical Center's twenty hospitals.
• How? When the Medical Center submitted a claim, it certified compliance with the Stark Act. The complaint makes all the allegations discussed above. We will not repeat them. But they detail exactly how these claims violated the Stark Act.
Rule 9(b) does not require the relators to plead anything more, such as the date, time, place, or content of every single allegedly false Medicare claim. The falsity here comes not from a particular misrepresentation, but from a set of circumstances that, if true, makes a whole set of claims at least prima facie false. It is enough to allege those circumstances with particularity. Doing so “inject[s] precision or some measure of substantiation into [the] fraud allegation” and “place[s] the defendant on notice of the precise misconduct with which [it is] charged.” Alpizar-Fallas v. Favero, 908 F.3d 910, 919 (3d Cir. 2018) (quoting Frederico v. Home Depot, 507 F.3d 188, 200 (3d Cir. 2007)) (last alteration in original; internal quotation marks omitted).
And the relators have done so. The second amended complaint runs 57 pages (plus exhibits) and comprises 257 numbered paragraphs. Dozens of these paragraphs go into great detail about specific physicians' Work Units and pay levels. The complaint compares those figures at length with industry benchmarks, medians, and 90th percentiles. It alleges specific ways that surgeons padded their bills, by for instance falsely reporting unperformed work assisting other surgeons or physically supervising residents and interns. The complaint also quotes the government's settlement agreement, alleging specific ways that surgeons had been padding their bills. The sum total of these allegations tells a detailed story about how the defendants designed a system to reward surgeons for creating and submitting false claims. See Omnicare, 903 F.3d at 91–92 (quoting Foglia, 754 F.3d at 158). And that is particular enough to satisfy Rule 9(b).
C. Pleading Stark Act exceptions under the False Claims Act
One final issue is how the Stark Act interacts with the False Claims Act. The defendants argue that the False Claims Act's elements of falsity and knowledge turn the Stark Act's exceptions into prima facie elements of the False Claims Act. On their reading, the relators would have to plead that no exception applies here.
We reject that argument. The defendants retain the burden of pleading Stark Act exceptions even under the False Claims Act. And even if the relators bore that burden, they have met it here.
1. The burden of pleading Stark Act exceptions stays with the defendant under the False Claims Act. The defendants argue that the False Claims Act's knowledge and falsity elements turn the Start Act's exceptions into prima facie elements. Their logic is simple and cogent: The False Claims Act penalizes only false claims. 31 U.S.C. § 3729(a)(1). False claims include claims submitted in violation of the Stark Act. See Kosenske, 554 F.3d at 94. But if an exception to the Stark Act applies, then the claim is not false. And if the defendant thinks that an exception applies, then the defendant does not know that the claim is false. So, according to the defendants, to plead a False Claims Act claim based on Stark Act violations, a relator must plead that no Stark Act exception applies and that the defendant knows that none applies. Otherwise, the relator pleads neither falsity nor knowledge.
Though this argument has force, we reject it. Our precedent compels this result. Like this case, Kosenske was a False Claims Act case based on Stark Act violations. Id. It placed the burden of proving a Stark Act exception on the defendant. Id. at 95; accord Tuomey, 792 F.3d at 374. And we see no reason to split up the burdens of pleading and persuasion. It is thus the defendants' burden to plead a Stark Act exception, not the relators' burden to plead that none exists.
2. Even if the relators bore this pleading burden, they have met it. In any event, the relators here plausibly plead that no Stark Act exception applies. The parties identify four that could apply here: exceptions for bona fide employment, personal services, fair-market-value pay, and indirect compensation. All four exceptions require that the surgeons' compensation not exceed fair market value and not take into account the volume or value of referrals.
We have already explained how the relators plausibly plead that the surgeons were paid more than fair market value. And that itself suggests that their pay may take into account their referrals' volume or value. So the relators plausibly plead that no Stark Act exception applies.
Evaluating a motion to dismiss is “a context-specific task that requires the reviewing court to draw on its judicial experience and common sense.” Iqbal, 556 U.S. at 679, 129 S.Ct. 1937. Our experience and common sense tell us that the relators state a plausible claim that the Medical Center and Pittsburgh Physicians have violated the Stark Act and the False Claims Act.
The facts they plead, if true, satisfy every element of those statutes: A chain of financial relationships linked the hospitals to the surgeons. The surgeons referred many designated health services to the hospitals, generating ancillary hospital services and facility fees. It is plausible that their pay takes into account the volume of those referrals. The hospitals made Medicare claims for those referrals. And the defendants allegedly knew all this.
With all this smoke, a fire is plausible. So this case deserves to go to discovery. Once the discovery is in, it may turn out that there is no fire. We do not prejudge the merits. But this is exactly the kind of situation on which the Stark and False Claims Acts seek to shed light. We will thus reverse the District Court's dismissal and remand for further proceedings.
BIBAS, Circuit Judge.