IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF PENNSYLVANIA
UNITED STATES OF AMERICA, THE : CIVIL ACTION STATE OF MARYLAND, MICHAEL : GOEBEL, Ex Rel., and WILLIAM : COLEMAN, Ex Rel. : : v. : : SELECT REHABILITATION, INC., : ANCHORAGE SNF, LLC, d/b/a : Anchorage Healthcare Center, : COMMUNICARE HEALTH SERVICES, : INC. and WHITE OAK HEALTHCARE, : LLC d/b/a White Oak Senior Care : NO. 19-3277
MEMORANDUM OPINION
Savage, J. September 29, 2023
This is the second of four qui tam actions filed under the False Claims Act (“FCA”) against Select Rehabilitation, Inc. (“Select”)1 and operators of skilled nursing facilities (“SNFs”) alleging they billed Medicare and Medicaid for therapies that patients did not need or were not provided.2 All four actions name Select as a defendant, and each names different SNFs. We must decide whether this action is barred by the first-to-file rule codified in 31 U.S.C. § 3730(b)(5). The issue is whether the adding of new defendants in a second qui tam action takes the case outside the reach of the first-to-file rule. We hold that an action naming unrelated defendants not named in the first action, which does not sufficiently allege a nationwide fraudulent scheme, is not barred by the first-to-file rule. This is especially true here where the two actions show there were multiple conspiracies with Select at the center of each. Relators Michael Goebel and William Coleman (“Relators”) brought this action on March 16, 2017 in the United States District Court for the District of Maryland. On July 24, 2019, the case was transferred to this court where the case of United States ex rel. Patrick Gerard Carson v. Select Rehabilitation, Inc., No. 15-5708, which had been filed on October 20, 2015 (“Carson action”), was pending.3 The cases remained under seal
while the government investigated the claims. On July 5, 2022, the government declined to intervene. With the government’s consent, Relators voluntarily dismissed Select without prejudice on December 28, 2022, leaving only the SNF defendants. Moving to dismiss the Complaint, defendants CommuniCare Health Services, Inc., White Oak Healthcare, LLC, and Anchorage SNF, LLC (collectively “CommuniCare”) invoke the FCA’s first-to-file bar. They contend that this action is based on the same essential facts as those alleged in the Carson action, which was filed first. Relators counter that this case involves a different fraudulent scheme between different
defendants at different facilities in different states. This action and the Carson action each describe a fraudulent scheme involving Select and SNFs to maximize reimbursements from Medicare and Medicaid by overbilling for therapy services; billing for therapy services that were not provided; manipulating billing practices; and billing for medically unreasonable and unnecessary therapy. This action includes new unrelated defendants whom we cannot conclude the government would have discovered in its investigation instigated by the Carson action. The two actions allege similar, but separate, schemes with Select central to both. Thus, we hold that this action is not barred by the first-to-file rule. Analysis Section 3730(b)(5) provides: “When a person brings an action under this subsection, no person other than the Government may intervene or bring a related action based on the facts underlying the pending action.” 31 U.S.C. § 3730(b)(5). This first- to-file rule bars a later related FCA action. United States ex rel. St. John LaCorte v.
SmithKline Beecham Clinical Labs., Inc., 149 F.3d 227, 232 (3d Cir. 1998). An action is related if it alleges the same essential facts of the claims in the prior action. Id. at 232–33. A complaint adding new facts and providing more details of the same claim alleged in the earlier action is barred by the first-to-file rule. The later case need not rest on the exact or identical facts as the prior case to fall within the bar. “Rather, if a later allegation states all the essential facts of a previously-filed claim, the two are related and section 3730(b)(5) bars the later claim, even if that claim incorporates somewhat different details.” Id. The dispositive question is: does the later case repeat causes of action based on the same fraudulent scheme.
To determine whether the actions are related necessarily requires that we compare the two complaints. In doing so, we conduct a claim-by-claim analysis. United States ex rel. Merena v. SmithKline Beecham Corp., 205 F.3d 97, 102 (3d Cir. 2000). Relator Patrick Gerard Carson, a physical therapist assistant employed by Select from October 2011 to March 2015, brought the first-filed action.4 He asserts violations under sections 3729(a)(1)(A)–(G), 3730(h) of the FCA, and 15 states’ false claims laws. Those states are Connecticut, Delaware, Florida, Georgia, Illinois, Indiana, Maryland, Massachusetts, New Jersey, New York, North Carolina, Rhode Island, Texas, Virginia, and Washington. As Relators in this action do, he named Select as the central defendant. He also named five skilled nursing facilities and the companies that owned them.5 All the SNFs are in Pennsylvania.6 None are named in this action. The Carson complaint alleges that Select and the SNF defendants engaged in a fraudulent scheme, orchestrated by Select, to submit false and fraudulent claims to Medicare and Medicaid.7 The scheme was designed to bill the maximum per diem
reimbursement rate for patients receiving therapy regardless of need and for therapy that was not provided.8 As described in the Carson Complaint, the scheme worked as follows. Select contracted with SNFs to staff them with physical and occupational therapists and assistants to deliver rehabilitation therapy services.9 The therapists and assistants followed Select’s policies, delivering occupational and physical therapies to patients who did not need them, could not participate in them, or would not benefit from them.10 At management’s direction, therapists and therapist assistants employed or contracted by defendants manipulated medical and billing records to falsely inflate therapy provided and report therapy not provided.11 Based on these records, the SNFs submitted false and
fraudulent bills to Medicare and Medicaid. Carson asserts the defendants instructed or encouraged the rehabilitation staff to: (1) shift treatment minutes among the three therapy disciplines (physical, occupational and speech), regardless of medical need; (2) improperly manipulate use of concurrent therapy; (3) reduce the amount of therapy provided to privately insured patients; (4) overutilize modalities, such as electric stimulation and ultrasound; (5) utilize inappropriate modalities; (6) improperly bill for treatment; and (7) bill for services not provided.12 Defendants also postponed the discharge of patients for days and sometimes weeks longer than medically appropriate to increase reimbursements.13 As described by Carson, Select classified patients in the highest possible resource utilization group (“RUG”) so it could bill Medicare and Medicaid at the highest per diem reimbursement rate regardless of the need and whether the therapy was provided.14
For Medicare purposes, patients are categorized into RUG levels.15 The RUG levels are: • Ultra-High (a minimum of 720 minutes per week on at least five distinct days);
• Very-High (a minimum of 500 minutes per week on at least five distinct days);
• High (a minimum of 325 minutes per week on at least five distinct days);
• Medium (a minimum of 150 minutes per week on at least five distinct days); and
• Low (a minimum of 45 minutes per week on at least five distinct days).16
The higher the number of therapy minutes or the higher RUG level, the higher the reimbursement.17 Carson claimed that there was “constant pressure to capture the highest number of treatment minutes for Government-funded health programs” that “often resulted in circumstances that put the health and welfare of the patients at risk” as “[p]atients were often left unattended and forced to engage in training without proper support and supervision.”18 Carson listed numerous examples of patient cases to illustrate how the defendants inflated time reports used to bill Medicare and Medicaid.19 Carson describes a corporate culture that encouraged “obtain[ing] the highest reimbursement possible for skilled nursing facility stays and the therapy administered during those stays.”20 Select achieved its billing goals through policies of “maximizing productivity rates of its therapists”21 and “retaining patients based on billing potential rather than patient well-being.”22 Carson added that there was constant pressure from management to focus “on obtaining high RUG categories . . . [and] captur[ing] the highest number of treatment minutes,”23 even when patients did not receive or need the treatment. In short, Select expected therapists to bill as much as possible for as long as
possible, without regard to medical need. Relators Michael Goebel and William Coleman brought this action, alleging violations of the False Claims Act, 31 U.S.C. §§ 3729(a)(1)(A)–(C), and (G), and the Maryland False Claims Act Against State Health Program Act, Md. Code Ann., Health- Gen. § 2-601, et seq. As Carson does, they named Select as the central defendant.24 Select has since been voluntarily dismissed.25 They also name one skilled nursing home, Anchorage, and the two companies that owned it at different times, CommuniCare and White Oak.26 Carson did not name these defendants. Goebel, a certified occupational therapy assistant licensed in the State of Maryland, has worked for Select since September 2010.27 He was the Program Manager
at Anchorage for five years, providing health care services, evaluating patient functional outcomes, maintaining patient resource utilization group levels, scheduling patient caseloads, managing therapists and assistants, and attending meetings to ensure safe and timely discharge of patients.28 William Coleman, a physical therapist licensed in the State of Maryland, worked for Select from July 2015 until his resignation on September 9, 2016.29 Anchorage was one of the facilities where he worked.30 As does the Carson complaint, the Goebel complaint alleges a scheme to submit false and fraudulent claims to Medicare and Medicaid to maximize reimbursements.31 Like Carson, Goebel alleges that Select staffed SNFs with therapists and assistants who carried out Select’s policy of overbilling Medicare and Medicaid patients.32 Select controlled the type and scope of therapy provided and billed at Anchorage,33 which
submitted the false and fraudulent bills to Medicare and Medicaid for reimbursement. CommuniCare knew about Select’s practices but allowed the scheme to continue because it profited from it.34 Relators detail a four-part scheme designed to maximize reimbursement from Government-funded programs. First, “Select routinely provides skilled care for patients based on illegitimate, medically irrelevant, falsified, and non-CMS compliant standards.”35 In other words, Select provided therapies to patients that did not meet eligibility standards and did not qualify for Medicare reimbursement. Decisions about patients’ medical care were made by off-site Select administrators who were “under enormous corporate
pressure to secure and report high RUG levels for each patient” regardless of medical necessity.36 Administrators “routinely disregard doctor’s orders and override recommendations and suggestions of licensed, treating therapists.”37 Second, “Select knowingly falsifies medical records and patient assessment forms submitted to its fiscal intermediary and to CMS, in order to up-code[38] RUG levels. . . and improperly increase therapy.”39 The Goebel complaint provides examples of how Select carried out the scheme by submitting false claims for services that had been intentionally up-coded;40 falsifying patient reports and fabricating patient evaluations, plans of treatment, and therapy progress reports;41 and ignoring patient tolerance for therapy to classify all patients in the Ultra-High RUG level.42 Upcoding resulted in providing therapy that was medically unreasonable, unnecessary and potentially harmful.43 Third, Select engaged “in strategic methods to continue its fraudulent practice undetected.”44 It had a practice of “classic ramping.” It “is the practice of setting treatment minutes at an Ultra-High level right before the assessment reference date and right before
the assessment cycle ends, only ramping down the treatment minutes when the new assessment cycle begins.”45 Fourth, Select did not staff therapists based on patient demand.46 It staffed therapists and clinicians and then built a patient caseload to match staffing availability.47 Its corporate directives promoted maximizing profit over patients’ needs.48 As a result, patients received the highest level of therapy whether they needed it or not.49 Goebel alleges Select’s regional administrators, regional managers, and program managers enforced corporate management orders at Anchorage.50 These managers, working off-site, made clinical decisions about patients’ therapy treatments at Anchorage,
including decisions about billing, RUG levels, and treatment minutes without regard to patient need.51 As an incentive to maximize Medicare reimbursement, Select awarded administrators and managers financial bonuses.52 Because it staffed Anchorage with physical and occupational therapists, Select was able to control the amount of therapy provided to patients and billed to Medicare and Medicaid.53 Select also hired therapy assistants that were easier to control and less likely to question management.54 The therapy assistants followed management’s recommendations for patient treatment minutes without question.55 Relators allege that CommuniCare and White Oak conspired with Select to submit false and fraudulent claims to Medicare and Medicaid.56 CommuniCare and White Oak allowed Select to control therapies at Anchorage and helped implement Select’s corporate directives.57 In return for their participation, CommuniCare and White Oak received a portion of the reimbursements.58 Because they were beneficiaries of the
scheme, they acquiesced in the fraud.59 Relators detail how Select controlled the therapy programs and set the therapy minutes for each patient without regard to actual need. Bobby Schaffer, Select’s Northeast Regional Manager who supervised Relators, set the RUG levels and assigned treatment minutes at Anchorage.60 Schaffer worked off-site where he set unrealistic daily percentages of RUG levels and assigned RUG levels to patients without appropriately evaluating them.61 In assigning RUG levels, Schaffer ignored on-site documented patient needs, overriding doctors’ orders and therapists’ treatment notes.62 Schaffer micromanaged and controlled patient discipline schedules and ordered the clinical staff to assign the highest level of therapy to patients without examining them.63
Schaffer’s supervisor Ed Luberski, Regional Vice President at Select, established a case management index policy for Anchorage.64 As Luberski directed him, Schaffer set the daily discipline schedule at Anchorage.65 Therapists could not start working until they received the daily minutes from Schaffer.66 The CommuniCare defendants argue that the Goebel complaint alleges the same essential facts as the pending first-filed Carson action. They acknowledge that the complaint in this action contains additional details, new defendants and different time periods. But, they contend, these differences are immaterial and that what matters is that the complaint alleges the same essential facts and the same fraudulent scheme. Relators counter that this action alleges not only different facts and different defendants, but a different fraudulent scheme. Specifically, they contend that the Carson action involves “a separate whistleblower in a separate case [that] had previously sued
separate defendants in a separate state for separate fraudulent schemes related to Medicare and Medicaid reimbursement.”67 Despite the apparent similarities of the complaints, the Relators contend that the complaints allege different fraudulent schemes. They cite five examples of how the fraud that occurred at Anchorage differs from the fraud described in Carson. These allegations include the falsification of minimum data set assessments and service log matrices to manipulate RUG levels; pressuring clinicians to meet Medicare Part B patient projections and increasing the length of patient stay irrespective of need; ramping up and down therapy minutes during assessment reference periods to maximize reimbursements; and
enforcing corporate directives that encourage therapists to maintain unattainable productivity levels. That the facts are not identical is of no moment. The conduct Relators characterize as differences are additional details more fully describing a similar scheme as set forth in Carson. They describe essentially the same facts central to the fraudulent schemes. There is no substantial difference between the fraudulent schemes alleged in Carson and Goebel. Adding facts and details is not enough to avoid the first-to-file bar. The rule does not require the facts in the later action be identical to those in the pending action to trigger the statutory bar. LaCorte, 149 F.3d at 234. As the First Circuit emphasized: We have made clear the first-to-file rule does not necessarily protect more detailed, later-filed complaints from less detailed, earlier-filed ones. So long as the first complaint sets forth the “essential facts” of the fraud alleged in the second complaint, it does all it needs to do under the first-to-file rule.
United States ex rel. Ven-A-Care of the Fla. Keys, Inc. v. Baxter Healthcare Corp., 772 F.3d 932, 939 (1st Cir. 2014) (internal citation omitted); see also United States ex rel. Wood v. Allegran, Inc., 899 F.3d 163, 169 (2d Cir. 2018) (“Though [relator’s] allegations may be more detailed than those asserted in [the first case], the two cases in essence alleged very similar kickback schemes.”). Here, the essential facts are materially similar, albeit presented differently and more explicitly. Relators argue that the first-to-file bar does not apply because this case was brought by different whistleblowers who worked at different facilities in different states at different times. These differences are not material. A relator cannot overcome the first- to-file bar by adding “geographic locations to the essential or material elements of a fraud claim.” United States ex rel. Branch Consultants v. Allstate Ins. Co., 560 F.3d 371, 378 (5th Cir. 2009). Nor does working for Select at different times overcome the bar. See United States ex rel. Poteet v. Medtronic, Inc., 552 F.3d 503, 517–18 (6th Cir. 2009) (citations omitted). Relators also contend that the cases are not related because they name different defendants. The CommuniCare defendants counter that this difference is immaterial. What matters, they say, is that the complaint alleges the same essential facts and the same fraudulent scheme. Carson named five SNFs and three SNF owners. All the SNFs were in Pennsylvania. Goebel names one Maryland SNF and its current and former owners. There is no overlap in SNFs or their owners. However, there is one common defendant— Select, the central defendant in both actions. Relators insist there is no overlap of defendants. That is not correct. At the time
this action was filed, Select was named in both actions. Although Relators named different SNFs as defendants, they named Select as the defendant at the center of the fraudulent scheme. Only after the seal was lifted and the complaint was served did the Relators voluntarily dismiss Select. Does the adding of new defendants in another state remove the case from the reach of the first-to-file bar? The answer to this question resides in the answers to two questions—does the later action allege the same fraudulent scheme as the first action and did the first complaint give the government enough information to identify the defendants added in the later-filed complaint as participants in the scheme.
We begin with examining how different circuits determine whether qui tam cases are related. Then, we look at how these courts apply their relatedness tests to later actions naming new defendants. In determining whether a later case is related to the first-filed case, the First, Third, and Sixth Circuits apply an “essential facts” test. See, e.g., United States ex rel. Ven-A- Care of the Fla. Keys, Inc. v. Baxter Healthcare Corp., 772 F.3d 932, 939 (1st Cir. 2014); United States ex rel. St. John LaCorte v. SmithKline Beecham Clinical Labs., Inc., 149 F.3d 227, 232 (3d Cir. 1998); United States ex rel. Poteet v. Medtronic, Inc., 552 F.3d 503, 517 (6th Cir. 2009). The Fourth, Ninth, Tenth, and D.C. Circuits apply a “same material elements” test. See, e.g., United States ex rel. Carter v. Halliburton Co., 710 F.3d 171, 182 (4th Cir. 2013); United States ex rel. Lujan v. Hughes Aircraft Co., 243 F.3d 1181, 1189 (9th Cir. 2001); In re Natural Gas Royalties Qui Tam Litigation, 566 F.3d 956, 962 (10th Cir. 2009); United States ex rel. Hampton v. Columbia/HCA Healthcare Corp., 318 F.3d 214, 218 (D.C. Cir. 2003). The Seventh Circuit looks at whether the two cases
share the same “material facts.” See, e.g., United States ex rel. Chovanee v. Apria Healthcare Grp., Inc., 606 F.3d 361, 363 (7th Cir. 2010). In practice, the tests are the same. As the Fourth Circuit observed: In determining whether a complaint is similar enough as to be caught by the first-to-file bar, courts have applied variations of a common approach. Although the approaches vary, courts have almost uniformly rejected an “identical facts” test on the ground that the provision refers to a “related” action rather than an “identical” action. The courts also agree that differences in specifics—such as geographic location or added facts—will not save a subsequent case. The Third, Fifth, Sixth, Ninth, Tenth, and D.C. circuits have all adopted a “same material elements test.” Carter, 710 F.3d at 181 (citations omitted); see also Cho on behalf of States v. Surgery Partners, Inc., 30 F.4th 1035, 1042 (11th Cir. 2022) (quoting Wood, 899 F.3d at 169) (“Both parties suggest that we follow our sister circuits in adopting the ‘same material elements’ test, also called the ‘same essential elements’ test. . . . Under this test, two actions are related if they ‘incorporate the same material elements of fraud’. . . . That is, ‘to be related, the cases must rely on the same essential facts.’”). The Third Circuit has used both phrases in conducting its relatedness analysis. In LaCorte, the court held that “if a later allegation states all the essential facts of a previously-filed claim, the two are related and section 3730(b)(5) bars the later claim.” 149 F.3d at 232. The court proceeded to find that the “material elements of this claim are the same as those in LaCorte's Claim 1. . . . Accordingly, section 3730(b)(5) bars Claim 1.” Id. at 235. This language illustrates there is no difference in the two tests. How the circuit courts decide whether to apply the rule turns on whether they consider the identity of a defendant a material element or essential fact of the fraudulent scheme. If the new defendant is deemed material or essential, the bar does not apply.
Conversely, if the defendant is not considered a material element or essential fact, the later case is barred. The Third Circuit has not addressed whether actions based on the same fraud naming different defendants are related for purposes of applying the first-to-file rule. Other circuits have. The Tenth Circuit considers a defendant’s identity a material element of a fraud claim. Natural Gas, 566 F.3d at 962. Consequently, it held that “[t]wo complaints can allege the very same scheme to defraud the very same victim, but they are not the same claim unless they share common defendants.” Id. The Court explained:
Multiple parties can defraud the government through identical schemes. While we might consider a complaint that alleges an additional method of defrauding the government to state the same essential claim, we would not consider a complaint against an entirely different defendant to be stating the same claim. There is a difference between a relator who simply tacks on an additional piece of evidence . . . and a relator who alleges a scheme committed by a different party. The former might make it easier to prove a material element of the fraud and might even be the difference between success at trial or failure, but the latter asserts a different claim, seeking distinct damages arising out of a separate injury caused by another party.
Id.; see also United States ex rel. Lutz v. Berkeley Heartlab, Inc., 225 F. Supp. 3d 487, 507 (D.S.C. 2016) (“[A] later-filed action is not based on the facts of a pending action when it identifies a new defendant who is not a subsidiary of an already-named defendant.”). In some circuits, a defendant’s identity is not necessarily a material element of a fraud claim, but can be. The Sixth Circuit has held that “the fact that the later action names different or additional defendants is not dispositive as long as the two complaints
identify the same general fraudulent scheme.” Poteet, 552 F.3d at 517. Similarly, the Eleventh Circuit, while cautioning that there is no bright-line test to determine whether the addition of a new defendant creates a new claim, framed the question as “whether the introduction of a new defendant amounts to allegations of a ‘different’ or ‘more far- reaching scheme’ than was alleged in the earlier-filed action.” Cho, 30 F.4th at 2043 (citation omitted). If it does, the actions are not related. On the other hand, if the new defendants were part of the same fraudulent scheme, they are related and the later action is barred. In the District of Columbia Circuit, the critical factor is whether the two actions
allege the same scheme on a corporate-wide or nationwide basis, not the identities of the defendants. Hampton, 318 F.3d at 218. So, if the first action alleges a corporate-wide or nationwide fraud and the second action alleges the same fraudulent scheme with different defendants, the actions are related. If the first action does not allege a corporate-wide or nationwide scheme and the second action alleged the same scheme with different defendants, the actions are not related. Although the Hampton court held that the first-to-file rule barred a second action that added a new company and its employees as defendants because the different defendants were “not differences in the material elements of the fraud,” it did so based on the relationship of the defendants. Id. The new company was a subsidiary of the company named in the first action that alleged a corporate-wide scheme. Id. Likewise, the Tenth Circuit, in Grynberg v. Koch Gateway Pipeline Company, rejected relator’s argument that the first-to-file bar “cannot apply because he named as defendants some affiliated [defendant] entities that were not listed as defendants in the [first] suit” where
both “complaints alleged the same essential claim of fraud.” 390 F.3d 1276, 1280 n.4 (10th Cir. 2004). That is not the case here. The SNFs named in the Goebel action are not in the same corporate family as the defendants in the Carson action. They are not related. But, that does not necessarily mean the first-to-file bar does not apply. Take, for example, the Fifth Circuit’s Branch Consultants decision. 560 F.3d 371. The first-filed action in Branch Consultants named as defendants four insurance companies that participated in FEMA’s Write-Your-Own (“WYO”) flood insurance program.68 It alleged that they defrauded the government by mischaracterizing wind damage as flood damage
from Hurricane Katrina to shift the loss from the insurers to the government. Id. at 374. The complaint contained general allegations of fraud among the four WYO insurers and specific allegations against one of the WYO insurers. Id. at 374–75. The later-filed action named eight WYO insurers, including two that had been named in the first-filed action, and six adjusting firms. Id. The later action also alleged that the WYO insurers defrauded the government by the same means as the earlier action—mischaracterizing wind damage as flood damage. Id. at 375. Unlike the first-filed action, the later action went beyond general allegations of fraud, setting forth over fifty specific instances of fraud. Id. The district court in Branch Consultants had dismissed the later-filed action, “presumably on the theory that argued that [the first-filed action’s] broad allegations preempted the entire field of Katrina-related WYO fraud.” Id. at 379. The Fifth Circuit found that the first-filed action did “not allege a true industry-wide fraud or concerted action among a narrow group of participants.” Id. at 380. In holding that the first-to-file
bar did not apply, the Fifth Circuit reasoned that the first-filed action: implicates, at most, four specific WYO insurers among the approximately ninety-five WYO insurers conducting business in the Louisiana and Mississippi areas during Hurricane Katrina. Thus, [the first-filed action] tells the government nothing about which of the ninety-one other WYO insurers (and adjusting firms working for or with those insurers), if any, actually engaged in any fraud.
Id. The court pointed out that “unlike the additional defendants named in [other cases], the additional defendants named in this case are not corporate affiliates or subsidiaries of the . . . defendants.” Id. It explained that “there might be situations in which the allegations in a first-filed complaint pertain to such a narrow or readily-identifiable group of potential wrongdoers that § 3730(b)(5) acts to bar subsequent allegations against previously unnamed defendants.” Id. The test, it concluded, is whether the first action alleges enough facts to instigate a government investigation that would lead to the identity of the new defendants. Id. The court then determined that “here, nothing in the [first- filed] complaint provided the government with facts from which it could discern a widespread fraud involving all WYO insurers or the identities of other specific fraud– feasors.” Id. at 381. In the absence of Third Circuit precedent, we adopt the reasoning of the Fifth and Tenth Circuits and conclude that adding unrelated defendants in a later-filed FCA action does not necessarily bar that action under the first-to-file rule. We hold that a second action is not barred when it asserts a new claim based upon similar but different schemes and a separate injury caused by different defendants.
The question now is whether this action alleges different schemes. Both the Carson action and this action allege a scheme to defraud the government. The goal of the defendants in both actions was the same (to maximize reimbursement rates) and the means were the same (inflating and falsifying therapy records to support false bills). But, the players were different. A comparison of the complaints demonstrates that they allege separate conspiracies between a common defendant (Select) and each of the SNFs. There are multiple schemes alleged. The Supreme Court, in the criminal context, drew the distinction between a single
conspiracy and multiple conspiracies. It described a “rimless” conspiracy as one consisting of multiple conspiracies where there is a central defendant at the “hub” acting with several unrelated co-defendants—the “spokes.” Kotteakos v. United States, 328 U.S. 750, 753–54 (1946). The Court found that, absent some agreement or relationship between the spokes, there is no “rim of the wheel to enclose the spokes.” Id. at 755. Thus, in such a case, there is not a single conspiracy, but multiple conspiracies sharing a common defendant at the hub. The Kotteakos reasoning has been applied in the civil context. In Howard Hess Dental Lab'ys Inc. v. Dentsply Int'l, Inc., the Third Circuit held that in order to plead a single anti-trust conspiracy involving one central “hub” defendant, plaintiff must allege some agreement between the “spoke” defendants: Here, even assuming the Plaintiffs have adequately identified the hub (Dentsply) as well as the spokes (the Dealers), we conclude that the amended complaint lacks any allegation of an agreement among the Dealers themselves. . . . In other words, the “rim” connecting the various “spokes” is missing. 602 F.3d 237, 255 (3d Cir. 2010). Absent allegations suggesting a “unity of purpose, a common design and understanding, or a meeting of the minds” between and among the “spokes,” the plaintiff cannot establish the existence of a single conspiracy linking all the defendants. Id. at 257. The Third Circuit has applied the hub-and-spoke conspiracy principle to claims alleging a single enterprise in a civil RICO action. In In re Ins. Brokerage Antitrust Litig., the court affirmed the district court’s finding that “while plaintiffs may have alleged parallel, bilateral structures connecting a broker to each of its insurer-partners, they had failed to plead ‘broker-centered enterprises’ encompassing each broker ‘hub’ and all of its strategic partners.” 618 F.3d 300, 374 (3d Cir. 2010) (emphasis in original). Carson and Goebel have alleged distinct fraudulent schemes with Select at the “hub” and different SNFs and operating companies as the “spokes.” Neither action alleges any agreement or cooperative effort among the SNFs. There is nothing in the two actions linking the SNFs to each other. Even though the SNF’s had the same objectives and used the same methods, they acted independently of each other. The SNFs’ actions were not related and were not interdependent. One SNF’s conduct was not helpful to achieving the goals of the other SNFs. The SNF defendants have no connection with one another except Select’s involvement in the schemes. In short, comparing the allegations in both actions shows that Select was at the center of multiple separate schemes. This action fits in the Natural Gas holding. The Relators do not merely add new allegations of fraudulent conduct. They allege a separate, albeit a substantially similar or identical, scheme committed by different parties. As the Tenth Circuit describes the
difference, the later case “asserts a different claim, seeking distinct damages arising out of a separate injury caused by another party.” Natural Gas, 566 F.3d at 962. Having concluded that the two actions do not allege the same scheme, we consider whether the first action gave the government enough information that would have led it to the identities of the defendants in the later action. Whether a new defendant in a later action is material depends on the scope of the fraudulent scheme. The question is whether the government knows enough of the facts of the fraudulent scheme described in the first action to discover the identities of the new defendants. Stated differently, would the government’s investigation instigated by the
first action have led it to the identities of the defendants named in the later action? If so, the later action is barred. If not, the second action may allege a separate scheme that is not barred.69 Carson named five SNFs and three SNF owners. All the SNFs were in Pennsylvania. It also named Select, alleging Select “partners with 500 communities across 31 states.”70 It is those 500 facilities and their owners that may have been part of the Select fraudulent schemes. But, Carson informs the government nothing about which of the other 495 SNFs, if any, engaged in fraud. Instead, the allegations in Carson are limited to SNFs located in Pennsylvania. Carson sets forth allegations that Select orchestrated a scheme to submit false and fraudulent claims to Medicare and Medicaid designed to maximize the per diem reimbursement rates for therapy. Although Carson asserts claims under 15 state false claims laws, including Maryland, the complaint contains no allegations specific to those states. This action sets forth specific allegations of fraudulent conduct that took place in Maryland.
There are only two allegations in Carson’s Complaint that purport to allege a nationwide fraud. They are: Staff were told that employing modalities in the therapy regime made it seem more complex and skilled and would raise fewer questions from Medicare. In fact an individual specifically promoted and instructed the use of modalities. When certain staff raise concerns, the individual confirmed that many employees (across the country) complained that it was difficult to treat patients with modalities because management required that too many patients be treated at the same time.71
It was literally physically impossible to treat in this way. Thus, treatment was undertaken with no regard for whether the modalities were necessary, reasonable or beneficial to the patient. It also became apparent that this individual knew that these improper treatments were done on a nation-wide basis because of where he had worked.72
These paragraphs do not sufficiently allege that Select was carrying out a corporate-wide fraudulent scheme on a national basis. Instead, they allege that an unidentified individual claimed that “many employees (across the country)” complained it was difficult to treat patients with modalities and that improper modality treatments “were done on a nation-wide basis.” These facts do not make out a claim of fraud, let alone one on a national basis. Nor do they provide the government with enough information from which it could discern a company-wide fraudulent scheme between Select and the 495 other SNFs it supplied therapy services, including the Maryland SNF named in this action. We hold that naming unrelated defendants not named in Carson does not bar this action under the first-to-file rule. Here, the two actions identify separate fraudulent schemes orchestrated by the same defendant involving different players at different SNFs. Holding that adding unrelated defendants in a later-filed FCA action does not bar
the action under the first-to-file rule is not inconsistent with the rule’s purpose where the second action asserts a new claim based upon a separate injury caused by another unrelated party. One of the purposes of the rule is to prevent parasitic lawsuits that would diminish the recovery of the original relators which would have the effect of discouraging whistleblowers from coming forward. That is not a concern here. The Relators do not seek damages from Select or the Pennsylvania SNFs named in the Carson action. Instead, they pursue a Maryland SNF and its former and current owners. If they succeed, their recovery will not derive from Carson’s recovery. Conspiracy
The CommuniCare defendants move to dismiss Relator’s Complaint for failure to meet the particularity requirements under Federal Rules of Civil Procedure 8(a) and 9(b). They also argue that Relators fail to plead an underlying FCA violation. They add that Relators have not alleged a specific agreement between Select and the CommuniCare defendants to make out a conspiracy claim. Relators counter that the Complaint sufficiently alleges that Select conspired with CommuniCare and White Oak to violate sections 3729(a)(1)(A), (B) and (G) of the False Claims Act. According to them, there was an agreement between Select and the CommuniCare defendants to submit false or fraudulent Medicare and Medicaid claims by “inflating its therapeutic services on paper by various means so as to seek reimbursement from the government above and beyond what it was entitled to.”73 Relators add that the CommuniCare defendants acted in furtherance of the scheme “by ensuring that [Minimum Data Set] reports and [Resident Assessment Instruments] submitted to the government aligned with the false reports Select submitted.”74
Rule 9(b) requires that in “alleging fraud or mistake, a party must state with particularity the circumstances constituting fraud or mistake.” Fed. R. Civ. P. 9(b). To satisfy the particularity requirements of Rule 9(b), a complaint must state “the date, place or time of the fraud,” or otherwise inject “precision and some measure of substantiation into [the] allegations of fraud.” Lum v. Bank of Am., 361 F.3d 217, 224 (3d Cir. 2004) (quoting Seville Indus. Mach. Corp. v. Southmost Mach. Corp., 742 F.2d 786, 791 (3d Cir.1984)). The particularity requirement imposed by Rule 9(b) applies in FCA cases. United States ex rel. Schmidt v. Zimmer, Inc., 386 F.3d 235, 242 n. 9 (3d Cir. 2004). Relators have satisfied the Rule 8(a) plausibility and 9(b) particularity
requirements. Relators have sufficiently set forth the particularities—the who, where, when and how—of the false claims and the defendants’ role in them. They described who was involved in the scheme, where and when the scheme took place, and how each participant was involved. Relators’ Complaint describes with particularity a scheme involving Select, Anchorage, CommuniCare and White Oak to maximize reimbursements from Medicare and Medicaid by overbilling for therapy services provided; billing for therapy services that were not provided; manipulating billing practices; and billing for medically unreasonable and unnecessary therapy. Select staffed Anchorage with therapists and assistants, who inflated or manipulated therapy provided, resulting in overbilling Medicare and Medicaid. Select staffed Anchorage with therapists and assistants through whom Select controlled the amount of therapy provided to patients and billed to Medicare and Medicaid.75 False and fraudulent bills were then submitted to Medicare and Medicaid. CommuniCare, and formerly White Oak, participated in the scheme by allowing Select to control the therapies at Anchorage and helped implement
Select’s corporate directives.76 In exchange, CommuniCare and White Oak received a portion of the fraudulently billed services.77 The Relators state when they worked at Anchorage.78 These allegations are sufficient to put the defendants on notice of what they are accused—violating sections 3729(a)(1)(A), (B) and (G) of the FCA. Having determined that Relators satisfy Rules 8(a) and 9(b), we now address whether they have sufficiently stated a cause of action of conspiracy under section 3729(a)(1)(C). Section 3729(a)(1)(C) states that “any person who . . . conspires to commit a violation of subparagraph (A), (B), (D), (E), (F), or (G) . . . is liable to the United States
Government for a civil penalty of not less than $5,000 and not more than $10,000 . . . plus 3 times the amount of damages which the Government sustains because of the act of that person.” 31 U.S.C. § 3729(a)(1)(C). To state a cause of action for conspiracy under section 3729(a)(1)(C), Relators must allege “(1) a conspiracy to get a false or fraudulent claim allowed or paid and (2) an act in furtherance of the conspiracy.” U.S. ex rel. Atkinson v. PA. Shipbuilding Co., 473 F.3d 506, 514 (3d Cir. 2007). They must allege facts showing that there was an agreement among conspirators to commit a fraud. Relators must also plead an underlying violation of the FCA. See United States ex rel. Petras v. Simparel, Inc., 857 F.3d 497, 507 n.53 (3d Cir. 2017) (citing Pencheng Si v. Laogai Research Found., 71 F. Supp. 3d 73, 89 (D.D.C. 2014)). Based on the allegations in the Complaint and drawing all reasonable inferences in Relators’ favor, we conclude that they have stated a plausible cause of action for conspiracy. They allege that the CommuniCare defendants and Select conspired to
present false or fraudulent claims for reimbursement to Medicare and Medicaid.79 They contend that CommuniCare and White Oak, as owners and operators of Anchorage, were in “constant communication” and “worked closely” with Select to manage therapy services, including “setting therapy minutes, monitoring therapists, assigning service to Medicare patients, managing productivity levels, and enforcing corporate directives.”80 CommuniCare and White Oak joined in and benefitted from the fraud by receiving a portion of the fraudulently billed services.81 Anchorage played “a key role in contributing to the fraud by carrying out the corporate directives that directly [led] to patient harm and government fraud.”82 Together, so Relators contend, Select and the CommuniCare
defendants inflated therapy minutes to maximize reimbursements for the financial benefit of all defendants involved.83 These allegations are sufficient to show a conspiracy as well as underlying violations of the FCA. Relators’ Complaint alleges that Anchorage, CommuniCare, and White Oak committed underlying violations of the FCA. Relators allege that the CommuniCare and White Oak conspired with Select to (1) knowingly present or cause to be presented false or fraudulent claims for payment to the government in violation of 31 U.S.C. § 3729(a)(1)(A);84 (2) knowingly make, use, or cause to be made or used, false records or statements material to a false or fraudulent claim to the government in violation of 31 U.S.C. § 3729(a)(1)(B);85 and (3) knowingly conceal or avoid an obligation to repay money to the government in violation of 31 U.S.C. § 3729(a)(1)(G).86 As we have discussed, Anchorage knowingly submitted the false and fraudulent bills with CommuniCare’s and White Oak’s knowledge and agreement. Thus, we conclude they have stated a cause of action for conspiracy under 31 U.S.C. § 3729(a)(1)(C).
Conclusion This action alleges a similar fraudulent scheme as Carson does—Select, together with SNFs and their owners, submitted false and fraudulent claims to Medicare and Medicaid to maximize payments. This action identifies new unrelated defendants in another state that may not have been discovered by the government. Thus, because it adds materially to the Carson claim, the first-to-file rule does not bar this action against CommuniCare Health Services, Inc., White Oak Healthcare, LLC, and Anchorage SNF, LLC.
1 Select is a national company that contracts with over 600 medical facilities in 33 states to provide rehabilitation therapy, staffing them with physical therapists, occupational therapists, and registered nurses who deliver post-acute rehabilitative services, including speech, occupation, and physical therapies. Compl. & Demand for Jury Trial ¶¶ 4, 37, 120, 122, ECF No. 1 [“Goebel Compl.”]. 2 Three of the qui tam actions were filed in the Eastern District of Pennsylvania. See United States ex rel. Carson v. Select Rehabilitation, Inc., No. 2:15-cv-05708 (E.D. Pa. filed Oct. 20, 2015); United States ex rel. Rose v. Select Rehabilitation, Inc., No. 5:17-cv-05434 (E.D. Pa. filed Dec. 1, 2017); United States ex rel. Sheehan v. Select Rehabilitation, Inc., No. 2:19-cv-01317 (E.D. Pa. filed Mar. 29, 2019). 3 On April 4, 2023, pursuant to Md. Code Ann., Health-Gen. § 2-604(a)(7), the claims under Maryland law were dismissed without prejudice. Section 2-604(a)(7) states: “If the State does not elect to intervene and proceed with the action under paragraph (6) of this subsection, before unsealing the complaint, the court shall dismiss the action.” Md. Code Ann., Health-Gen. § 2-604(a)(7). 4 United States ex rel. Carson v. Select Rehabilitation, Inc., No. 2:15-cv-05708 (E.D. Pa. filed Oct. 15, 2015) [“Carson Compl.”]. 5 Id. ¶¶ 15–22. 6 Id. ¶¶ 15–16, 18, 20, 22. 7 Id. ¶¶ 4–7. 8 Id. ¶¶ 57–59. 9 Id. ¶¶ 13, 15–22. 10 Id. ¶¶ 5–6. 11 Id. 12 Id. ¶ 6. 13 Id. ¶ 7. 14 Id. ¶ 59. 15 Goebel Compl. ¶ 110; Carson Compl. ¶¶ 43–45, 52. 16 Goebel Compl. ¶ 111. 17 Id. ¶ 112–13. 18 Carson Compl. ¶ 59. 19 Id. ¶¶ 77–79, 83–85, 89–98, 104–05, 120–22, 132–33, 135–36. 20 Id. ¶ 57. 21 Id. ¶ 84. 22 Id. ¶ 131. 23 Id. ¶ 59. 24 Goebel Compl. ¶ 37. 25 Order (Dec. 28, 2022), ECF No. 88. 26 Goebel Compl. ¶¶ 5–7, 38–40. CommuniCare is a national provider of post-acute care. Id. ¶ 39. It manages and operates 51 facilities, including skilled nursing rehabilitation facilities, long-term facilities, assisted living communities, independent rehabilitation facilities, and long-term acute care hospitals in Ohio, Maryland, Pennsylvania, West Virginia, and Missouri. Id. ¶¶ 6, 39. One of the facilities owned and operated by CommuniCare is Anchorage, a private SNF in Salisbury, Maryland. Id. ¶¶ 5–6, 38–39. Anchorage specializes in short and long-term rehabilitation and senior health care services. Id. ¶ 5. It provides nursing and therapy care to its patients. Id. Anchorage contracts with outside therapy providers, like Select. Id. ¶ 6. As the Complaint explains, “CommuniCare loans the Anchorage facility to Select and in turn, Anchorage contracts with Select to staff its facility with therapists provided by Select.” Id. Select and CommuniCare work together “to operate and manage Anchorage on the therapy side of the business, including setting therapy minutes, monitoring therapists, assigning service to Medicare patients, managing productivity levels, and enforcing corporate directives.” Id. 27 Id. ¶ 35. 28 Id. 29 Id. ¶ 36. 30 Id. 31 Id. ¶¶ 9–32. 32 Id. ¶¶ 4–5, 11. 33 Id. ¶¶ 11, 13. 34 Id. ¶ 10. 35 Id. ¶ 12. 36 Id. ¶ 13. 37 Id. ¶ 19. 38 Upcoding is the practice of using a code with a higher reimbursement rate when it is actually a code having a lower rate. Here, patients were regularly upcoded to ultra-high therapy levels to maximize reimbursement, even if patients were not receiving ultra-high therapy. See id. ¶¶ 22–23, 30, 162, 164, 177, 187, 204–05, 225, 231, 236, 239–40, 242–43, 328. Upcoding is “where a nursing home or other provider inflate the cost of its bill to Medicare by claiming more intensive services were done than actually performed. In other cases, nursing homes provide treatments that were inappropriate.” Nursing Homes Said to Overbill U.S., The Wall Street Journal Online (Nov. 13, 2012). 39 Goebel Compl. ¶ 20. 40 Id. ¶ 9(a). 41 Id. ¶ 9(d). 42 Id. ¶ 9(e). 43 Id. ¶¶ 9(a), 9(b), 9(f). 44 Id. ¶ 25. 45 Id. ¶ 275. 46 Id. ¶ 133. 47 Id. 48 Id. ¶¶ 10–11, 127, 132. 49 Id. ¶ 126. 50 Id. ¶ 123. 51 Id. ¶¶ 124–26. 52 Id. ¶ 126. 53 Id. ¶¶ 130–31. 54 Id. ¶ 131. 55 Id. 56 Id. ¶¶ 10–11, 124–25. 57 Id. 58 Id. ¶¶ 124–25, 153. 59 Id. 60 Id. ¶ 128 61 Id. 62 Id. ¶ 129. 63 Id. 64 Id. ¶ 130. 65 Id. 66 Id. 67 Pls.’ Opp’n to Defs.’ Mot. to Dismiss the Compl. at 12, ECF No. 96 (emphasis in original) [“Pls.’ Opp’n”]. 68 The WYO program “allows private insurance companies to write and service, in their own names, responsible for determining the extent of an insured’s flood damage, which in turn determines the amount of benefit ultimately paid out by the Federal Treasury.” Branch Consultants, 560 F.3d at 374. This “program has rules applicable to all insurers in the program, [but] the program does not involve coordinated efforts by or joint cooperation among the participating insurers.” Id. 69 The government knows whether its investigation would have identified the CommuniCare defendants as involved in the alleged scheme. Although it declined to intervene, it retains its interest in the litigation and can intervene at any time. In this case, it had notice of the motions to dismiss. Yet, it did not participate. Consequently, we do not have the benefit of its disclosing whether it did or would have discovered the identities of the CommuniCare defendants during its investigation following the filing of the first-filed action. 70 Carson Compl. ¶ 13. 71 Id. ¶ 118. 72 Id. ¶ 119. 73 Pls.’ Opp’n at 34. 74 Id. 75 Goebel Compl. ¶¶ 5–6, 11, 18, 123-33, 369-70. 76 Id. ¶¶ 124–25, 369-70, 372, 374. 77 Id. ¶¶ 10–11, 125, 153. 78 Id. ¶¶ 35–36, 134. 79 Id. ¶ 367. 80 Id. ¶¶ 6–7, 125, 369–70. 81 Id. ¶¶ 10-11, 29, 372. 82 Id. ¶¶ 5, 10, 374. 83 Id. ¶¶ 10, 374. 84 Id. ¶¶ 368–70. 85 Id. ¶¶ 371–72. 86 Id. ¶¶ 373–74.