United States of America v. Curo Health Services Holdings, Inc.

UNITED STATES DISTRICT COURT

MIDDLE DISTRICT OF TENNESSEE

NASHVILLE DIVISION

UNITED STATES OF AMERICA ex rel. )

RACHEL CAROL ANDERSON and )

STEVEN TROY MATHIS, )

)

STATE OF TENNESSEE ex rel. )

RACHEL CAROL ANDERSON and )

STEVEN TROY MATHIS, )

)

Plaintiffs, )

)

v. ) Case No. 3:13-cv-00672 (Lead)

) Case No. 3:20-cv-00168 (Member)

CURO HEALTH SERVICES HOLDINGS, ) Judge Aleta A. Trauger

INC., CURO HEALTH SERVICES LLC )

f/k/a CURO HEALTH SERVICES, INC., )

TNMO HEALTHCARE, LLC d/b/a )

AVALON HOSPICE, and REGENCY )

HEALTH CARE GROUP LLC, )

)

Defendants. )

MEMORANDUM

Three motions to dismiss are pending in this case. First, the defendants have filed a

Renewed Motion to Dismiss the Consolidated Complaint in Intervention of the United States and

Tennessee (Doc. No. 147), to which the United States and Tennessee have filed a Response (Doc.

No. 153), and the defendants have filed a Reply (Doc. No. 163). Second, the defendants have filed

a Renewed Motion to Dismiss the First Amended Complaint of Plaintiff-Relators [Robin Dillon]

Teague, [Lisa] Pence, [Lisa] Adkins, and [Amy] Carnell (Doc. No. 148), to which those relator-

plaintiffs have filed a Response (Doc. No. 154), and the defendants have filed a Reply (Doc. No.

164). Finally, the defendants have filed a Motion to Dismiss the Third Amended Complaint of

Plaintiff-Relators Rachel Carol Anderson and Steven Troy Mathis (Doc. No. 165), to which those

relator-plaintiffs have filed a Response (Doc. No. 168), and the defendants have filed a Reply

(Doc. No. 170). For the reasons set out herein, the motion directed at the governments’ claims will

be denied, the motion directed at the claims of Teague et al. will be denied partially on the merits

and partially as moot, and the motion regarding the claims of Anderson and Mathis will be granted

in part and denied in part.

I. BACKGROUND1

In order to address the unique difficulties facing patients with terminal illnesses and their

loved ones, the medical profession has developed a model of care known as hospice, which is

designed, not to cure or even halt the patient’s disease, but to use medical technologies and

methods, as well as other supports, to ease the patient’s final days, months, or, in some cases, years

of life. See 42 C.F.R. § 418.3. But hospice is expensive, and many dying people cannot afford it

on their own. For certain qualifying patients, the costs of hospice are shouldered, at least in part,

by government healthcare programs such as Medicare and Medicaid.

This case is about a group of Tennessee hospice providers that, the plaintiffs allege, helped

themselves to a portion of the money available from those programs for hospice care by falsely

certifying that patients’ illnesses had reached a terminal stage, when, in fact, they had not. The

defendants do not argue—at least at this point—that the fraud claims against them should be

dismissed in their entirety. Rather, they ask the court to “narrow” the case in a number of ways,

based on the law governing how such causes of action should be pleaded. The defendants also ask

the court to dismiss a handful of additional claims, brought by former employees who helped bring

this situation to light, alleging retaliatory discharge.

1 There are three operative complaints in this case (Doc. Nos. 98, 100, 160), and, while these facts are

mostly taken from the governments’ Consolidated Complaint in Intervention (Doc. No. 100), which states

the central fraud allegations around which this case mostly revolves, each complaint’s allegations are taken

as true for the purposes of the motion to dismiss directed at that complaint’s claims.

A. Structure of Hospice Benefits Under Medicare and Medicaid

“To be eligible for hospice care under Medicare or Medicaid, a patient must be certified by

a physician as ‘terminally ill’—meaning that the patient’s prognosis ‘is for a life expectancy of 6

months or less if the terminal illness runs its normal course.’” U.S. ex rel. Holloway v. Heartland

Hospice, Inc., 960 F.3d 836, 842 (6th Cir. 2020) (quoting 42 C.F.R. §§ 418.20(b), 418.22(b)(1)).

The decision to admit a Medicare or Medicaid patient into hospice must be made on the

recommendation of the hospice provider’s “medical director in consultation with, or with input

from, the patient’s attending physician (if any).” 42 C.F.R. § 418.25(a). In making that decision,

the medical director must consider “(1) [d]iagnosis of the terminal condition of the patient[;] (2)

[o]ther health conditions, whether related or unrelated to the terminal condition[; and] (3) [c]urrent

clinically relevant information supporting all diagnoses.” 42 C.F.R. § 418.25(b). The

determination that an individual is hospice-eligible is documented and memorialized in a

document known as a “certification of terminal illness,” or “COTI,” which must initially be signed

by the patient’s “attending physician, if the individual has an attending physician,” as well as either

the medical director or a physician member of the hospice’s “interdisciplinary group,” or “IDG.”

42 C.F.R. § 418.22(c)(1). The IDG is a group of various types of professionals—including at least

one social worker and at least one pastor or counselor—“who work together to meet the physical,

medical, psychosocial, emotional, and spiritual needs of the hospice patients and families facing

terminal illness and bereavement.” 42 C.F.R. § 418.56(a)(1).

Some patients go into hospice with only a very short amount of time left. Other patients,

however, face a more uncertain calendar. Medicare and Medicaid measure the duration of hospice

placements in terms of “election periods,” so-called because they represent periods in which the

patient has elected to forego curative treatment in favor of palliative care. The initial election

period is 90 days; the second election period is another 90 days; and, after that, a qualifying patient

may remain in government-supported hospice for “[a]n unlimited number of subsequent 60–day

periods,” as long as his condition remains terminal, as defined by the relevant statutes and

regulations. 42 C.F.R. § 418.21. A fresh COTI must be completed at the beginning of each election

period, 42 C.F.R. § 418.22(a)(1), but, after the first COTI, only one physician’s signature is

required on each subsequent COTI, see 42 C.F.R. § 418.22(c)(2).

Although these policies originated in the Medicare program, Tennessee’s Medicaid

program, known as “TennCare,” follows them as well. See Tenn. Comp. Rules & Regs. 1200-13-

13-.04(1)(b)(11) (adopting “Medicare Hospice requirements”). (See also Doc. No. 100 ¶ 63 (citing

TennCare Policy Manual, Policy No. BEN 07-001 (Rev. 7)).) Under TennCare, however, hospice

eligibility has an additional level of significance in terms of a patient’s covered services, because

a valid COTI qualifies an eligible patient for residential nursing facility care, without that patient’s

having to go through TennCare’s ordinary Pre-Admission Screening and Resident Review process.

See Tenn. Comp. R. & Regs. 1200-13-01-.10(2)(c)(7), (i), (3); Tenn. Comp. R. & Regs. 1200-13-

01.-23(2)(d)(3).

If a patient’s illness progresses as originally expected—that is, if the patient dies before the

end of the six months concluding at the end of the second election period—then the patient’s COTI

can be based on a review of the patient’s health information, without an additional face-to-face

assessment. However, unless there is a public health emergency, “a hospice physician or hospice

nurse practitioner must have a face-to-face encounter with each hospice patient whose total stay

across all hospices is anticipated to reach the 3rd benefit period.” 42 C.F.R. § 418.22(a)(4)(i).

These face-to-face evaluations, which are performed for the purpose of allowing the practitioner

to “gather clinical findings to determine continued eligibility for hospice care,” must occur “no

more than 30 calendar days prior to . . . the 3rd benefit period recertification” and must be repeated

for “every benefit period recertification thereafter.” Id.

A COTI must meet the following requirements in order for Medicare or Medicaid to pay

for the patient’s hospice care:

(1) The certification must specify that the individual’s prognosis is for a life

expectancy of 6 months or less if the terminal illness runs its normal course.

(2) Clinical information and other documentation that support the medical

prognosis must accompany the certification. . . .

(3) The physician must include a brief narrative explanation of the clinical findings

that support[] a life expectancy of 6 months or less as part of the certification

and recertification forms, or as an addendum to the certification and

recertification forms. . . .

(4) The physician or nurse practitioner who performs the face-to-face encounter

with the patient [if required] must attest in writing that he or she had a face-to-

face encounter with the patient, including the date of that visit. The attestation

of the nurse practitioner or a non-certifying hospice physician shall state that

the clinical findings of that visit were provided to the certifying physician for

use in determining continued eligibility for hospice care.

(5) All certifications and recertifications must be signed and dated by the

physician(s), and must include the benefit period dates to which the

certification or recertification applies.

42 C.F.R. § 418.22(b); see also 42 U.S.C. § 1395f(a)(7). Pursuant to those requirements, the COTI

must not only contain certain information and be prepared in a certain way, see 42 C.F.R. §

418.22(b)(1), (3)–(5), but it must also be retained in a certain way—specifically, by inclusion in

the hospice agency’s own medical record for that patient, see 42 C.F.R. § 418.22(b)(2). Those

medical records, in turn, must themselves be “legible, clear, complete, and appropriately

authenticated,” or else the hospice can be found to be out of compliance with Medicare

requirements. 42 C.F.R. § 418.104. The effect of those requirements is that, if a hospice agency is

complying with Medicare and Medicaid rules, then either the agency or an authorized third party—

such as an auditor or investigator—should be able to refer to the record of each Medicare or

Medicaid patient in the agency’s care at any time and know immediately (1) the full and precise

basis of that patient’s eligibility for hospice; and (2) the important details of the patient’s care and

condition. (See Doc. No. 100 ¶ 57.)

Even the most detailed regimen of paperwork and recordkeeping, however, cannot change

the fact that it is often hard to predict how much time a person—even a very sick one—has left to

live. Medicare and Medicaid acknowledge the uncertainty of such predictions; that is why, even

though hospice care is generally reserved, under those programs, for individuals expected to live

only six or fewer more months, the programs expressly allow for such patients’ care to be

extended—and paid for—indefinitely, over the course of however many election periods are

required. There is, in short, nothing inherently wrong with a patient’s being in hospice care for

much longer than six months. Nevertheless, in an effort to ensure that the programs’ hospice

eligibility requirements are consistently, fairly, and accurately applied, the Center for Medicare

and Medicaid Services (“CMS”) has frequently provided guidance to assist physicians in

determining whether a patient’s prognosis falls within the six-month window required to meet the

definition of terminal illness. See, e.g., 79 Fed. Reg. 26538, 26556 (May 8, 2014); 78 Fed. Reg.

48234, 48247 (Aug. 7, 2013).

Among the sources that hospice agencies are expected to rely on are “local coverage

determinations,” or “LCDs,” propounded by Medicare Administrative Contractors, or “MACs,”

with whom CMS contracts on a regional basis to “process and make payments on valid claims.”

PHHC, LLC v. Azar, No. 1:18CV1824, 2018 WL 5754393, at *4 (N.D. Ohio Nov. 2, 2018). The

MACs to which the relevant agencies were assigned for the period relevant to this case—Cahaba

Government Benefit Administrator, LLC (“Cahaba”) and Palmetto GBA, LLC (“Palmetto”)—

issued LCDs “set[ting] forth clinical indicators for determining whether individuals with certain

diagnoses have life expectancies of six months or less.” (Doc. No. 100 ¶ 71.) Those LCDs

recognized that, while some diseases and conditions have a consistently short prognosis, other

diseases and conditions, such as Alzheimer’s disease, can be highly variable. For those diseases

and conditions, the LCDs identified certain factors relevant to expected life expectancy. (Id. ¶¶

73–74.)

The uncertainty of medical predictions presents a challenge for hospice admissions, but a

manageable one; the Medicare and Medicaid systems are capable of paying for a few, or even

many, extra months of hospice for beneficiaries whose diseases progress more slowly than had

been reasonably expected. An additional problem arises, however, when the difficulties of

prognosis are combined with the facts that (1) Medicare and Medicaid pay for hospice services on

a per diem basis and (2) the hospice agencies that certify hospice eligibility are the same parties

that receive those per diem payments. See 42 C.F.R. § 418.302. Paying hospice agencies per

diem—as opposed to, for example, paying them a one-time flat fee for each patient—means that,

the longer a patient is in hospice care, the more the hospice agency is paid.2 An incentive therefore

exists for unscrupulous providers to make a patient’s time in hospice as long as possible. Because

every hospice patient is dying, it is difficult to add days to the end of the hospice agency’s period

of care. It is far easier, rather, to add days to the beginning of it—which can be accomplished

simply by certifying the patient as terminally ill as early as possible. The inherent indeterminacy

of any given patient’s prognosis, moreover, makes it hard to detect when a hospice is stretching

2 The court stresses that the question of how best to pay for hospice care is a policy issue far beyond the

scope of this litigation or the court’s purview. It suffices to say that any payment structure is likely to lend

itself to some good practices and leave room for some abuses, and this case is about alleged behaviors that

were designed to exploit the current framework.

(or exceeding) the boundaries of reasonable medical evaluation in order to get patients into care—

and into the hospice agency’s paying population—sooner.

B. Allegations Against the Defendants

The defendants are all companies involved in providing hospice services in Tennessee.

TNMO Healthcare, LLC d/b/a Avalon Hospice (“Avalon”), which is the patient-level hospice

operator this case is about, owns and controls at least 27 hospice agencies in the state. (Doc. No.

100 ¶ 11.) Curo Health Services Holdings, Inc. (“Curo Holdings”) is a large operator of hospice

chains founded by a private equity firm. It has grown, in part, by purchasing smaller providers,

including, in 2011, Avalon’s then-parent company, Regency Healthcare Group, LLC (“Regency”).

(Id. ¶¶ 12–14.) Curo Holdings’ revenue is generated overwhelmingly—the governments estimate

about 95%—by Medicare. (Id. ¶ 14.) Curo Health Services, LLC (“Curo Health Services”) is a

subsidiary of Curo Holdings that, during the time period relevant to this case, shared officers and

a physical address with Curo Holdings. (Id. ¶ 22.) The court will refer to Curo Holdings and Curo

Health Services collectively as “Curo.”

The plaintiffs allege that Avalon frequently certified patients as hospice-eligible when it

should not have. For example, on June 29, 2010, a patient whom the Complaint in Intervention

refers to as Patient No. 1 was admitted to Avalon’s hospice agency in Jackson, Tennessee with a

principal diagnosis of purportedly “end stage” heart failure, typically referred to as “congestive

heart failure” or “CHF.” (Id. ¶¶ 210–11.) Heart failure can be deadly, but not necessarily

imminently so; whether Patient #1 was actually hospice-eligible therefore depended on an

evaluation of his particular condition to determine whether it supported a six-month prognosis. At

the time, Avalon Jackson was led by Director of Operations (“DOO”) Barbara Gordon, who,

according to the plaintiffs, regularly “took steps to circumvent clinicians’ concerns about hospice

eligibility that led to the admission and retention of ineligible patients.” (Id. ¶ 155.) With regard

to Patient No. 1, this included “instruct[ing] a former Avalon Jackson nurse to document that

Patient No. 1 had shortness of breath,” despite the fact that the nurse had observed no such

symptom. (Id. ¶ 213.) The plaintiffs also allege that Gordon “instructed nurses to . . .deduct five

pounds from the patient’s weight for ‘dry weight’” and take other steps to “exaggerate the extent

of the patient’s illness.” (Id. ¶ 213.)

Patient No. 1 lived another seven and one-half years, eventually dying on January 2, 2018.

(Id. ¶ 224.) The patient spent the majority of that time receiving hospice care from Avalon. Several

times during that period, Avalon and/or Curo personnel noticed that the patient’s health did not

appear to be declining. Indeed, Avalon began planning to discharge the patient as non-terminal in

2011. Instead, an Avalon medical director simply reclassified Patient No. 1 as suffering from a

different terminal illness—end stage dementia—and discontinued the discharge process. (Id. ¶¶

216–18.) But Patient No. 1 was not near death for that reason either. Patient No. 1 continued to

survive, and red flags about the patient’s prognosis continued to arise, until a June 2014 review by

a Curo clinical compliance monitor found enough unequivocal flaws in the documentation related

to Patient No. 1—including a claim that the patient suffered from “worsening MS” despite not

having a “previous diagnosis or documentation of MS”—that Patient No. 1 was finally discharged,

albeit only many months later, in late March of 2015. (Id ¶¶ 223–24.) Patient No. 1’s time in

Avalon’s hospice care spanned numerous election periods, resulting in dozens of Medicare claims,

most of them for thousands of dollars. (Id. ¶ 225.)

The plaintiffs provide other examples. Patient No. 2 was admitted to Avalon Jackson’s

hospice for coronary artery disease that, like Patient No. 1’s heart failure, had been exaggerated at

DOO Gordon’s direction. Indeed, according to the plaintiffs, DOO Gordon had to overrule two

nurses before she found one willing to support the admission. (Id. ¶¶ 231–34.) A few months later,

Patient No. 2 left hospice, sought treatment, and lived another three and one-half years. (Id. ¶¶

235–39.) According to the plaintiffs, Avalon had not even submitted its Medicare claims for

Patient No. 2 when it became clear that the patient was not terminal, but Avalon went on to submit

the claims—amounting to thousands of dollars—anyway. (Id. ¶ 242.) Patient No. 3 is another

patient who came to Avalon Jackson, received hospice services paid for by Medicare, was

discharged, and lived for multiple years thereafter. (Id. ¶¶ 244–58.)

Not all of the plaintiffs’ examples were from Avalon Jackson. Patient No. 4 was treated at

Avalon Tullahoma and received hospice services—some of them paid for by TennCare and some

by Medicare—for over five years. (Id. ¶ 301.) Patient No. 4 was admitted for ostensibly end stage

cirrhosis of the liver, despite the fact that quantitative testing suggested that his disease was not

yet at that point in its progression. (Id. ¶ 303.) As with Patient No. 1, Curo and/or Avalon personnel

raised concerns about Patient No. 4 multiple times, but the hospice care—and the billing—

continued until, many thousands of dollars later, Patient No. 4 was discharged. (Id. ¶ 322–24.)

The final exemplary patient identified, Patient No. 5, was a patient of Avalon Athens for

nearly six years, and the plaintiffs have described a similar pattern of numerous internal concerns

being dismissed until Patient No. 5 was finally discharged, having allowed Avalon to bill

thousands of dollars to government healthcare programs—in this instance, Medicare—which

Avalon did not return. (Id. ¶¶ 348–69.) Although the plaintiffs do not provide examples from every

Avalon agency in Tennessee, they assert that those other facilities had similar problems. In support

of that allegation, the plaintiffs state that former employees from other Avalon agencies, such as

Avalon Chattanooga and Avalon Johnson City, voiced concerns about Avalon receiving payments

for ineligible patients. (Id. ¶¶ 370–75.)

The pattern of admitting ineligible patients to Avalon facilities, the plaintiffs allege, was

not merely the result of scattershot errors or a few bad actors. Rather, “Regency and Curo pressured

Avalon staff to maximize census through aggressive targets, financial incentives, and restrictions

on discharging patients,” and Avalon’s ground-level employees simply responded, predictably and

reliably, to those top-down pressures. (Id. at 22.) The parent companies “set aggressive targets for

the number of hospice patient admissions” and “pressured Avalon staff to meet these high

admissions and census targets by tracking financial goals on a monthly, quarterly, and annual

basis.” (Id. ¶¶ 88–89.) Curo kept a monthly “scorecard” of each hospice agency’s average daily

census, referrals, admissions, referral conversion rate, revenues, and profits, and, if an agency

failed to meet Curo’s targets, the agency’s DOO was required to submit a 30-day action plan. (Id.

¶ 90.) Certain targets, when missed, triggered a process through which Curo scrutinized each

patient from the relevant period whom the hospice elected not to admit. As part of this review of

so-called “Not Taken Under Care” or “NTUC” cases, the agency was required to justify every

decision not to admit a patient to hospice. Some regional managers went so far as to require some

DOOs to notify them of every patient not admitted. (Id. ¶ 93.)

The pressure from Regency and, later, Curo was backed up by financial incentives. For

example, “hospice care coordinators”—effectively, marketers—were paid bonuses for admissions

in excess of the relevant facility’s target. (Id. ¶¶ 89, 95–99.) DOOs were given raises or paid

bonuses based on the size of their patient populations. (Id. ¶ 100.) Those bonuses could be

substantial; for example, the DOO of Avalon Nashville was offered a “monetization event bonus”

equivalent to 20% of her base salary if her census exceeded a certain number by a particular date.

(Id. ¶ 101.) The structure of the financial incentives and management cudgels changed over time

and after Avalon was taken over by Curo. They continued, however, to reward admissions and/or

patient census numbers in such a way that particular agency personnel stood to benefit financially

from admitting more patients, while effectively punishing the management of facilities that

admitted fewer patients. (Id. ¶¶ 101–15.)

The plaintiffs have alleged other ways that corporate policies enabled—and concealed—

overly aggressive admissions. For example, Curo trained personnel not to use words or phrases

that could undermine a terminal prognosis, such as “stable,” “no change,” “feels good,” or “looks

good.” (Id. ¶ 122.) A Curo-distributed “reference guide” for physicians instructed them not to

document any belief that a patient had a life expectancy greater than six months, because such

statements were “not helpful.” Instead, physicians were, at most, supposed to state, vaguely, that

such patients’ eligibility was under continued evaluation. (Id. ¶ 123.)

Curo frequently audited patient records and identified those for whom hospice admission

was not, based on those records, supported. The purpose of those audits, however, was, at least in

large part, not to get patients out of hospice and back into curative or life-extending care because

their admission had been in error or they had unexpectedly improved. Rather, the audits identified

opportunities to supplement the records of patients so that, if Medicare or Medicaid scrutinized

the record, the patient would be found to be hospice-eligible. (Id. ¶¶ 124–25.)

The result of these corporate policies was that Avalon agencies found ways big and small

to admit patients even if the patients’ eligibility was unsupported. For example, DOOs took steps

like bypassing patients’ primary care physicians who did not consider them terminal by recruiting

replacement physicians to refer the patients regardless. In at least one instance, Avalon even

allegedly “used [a] physician’s log-in credentials to complete this documentation without his

authorization.” (Id. ¶ 139.)

According to the plaintiffs, Regency and Curo were aware of these practices, because they

received internal complaints about them. Nevertheless, even when Regency or Curo corroborated

a complaint, it took insufficient steps to actually identify improperly admitted patients. Moreover,

as the plaintiffs’ examples attest, even in the cases in which a patient’s ineligibility was so apparent

that the plaintiff was discharged, Regency and Curo generally did not take the steps necessary to

identify and return overpayments. (Id. ¶ 140.) For example, in February of 2012, Curo’s

compliance hotline received a tip regarding falsification of documentation and admission of

ineligible patients at Avalon Jackson. (Id. ¶ 195.) Curo’s compliance personnel investigated, and

DOO Gordon quickly resigned. (Id. ¶ 198.) However, after Curo Vice President of Clinical

Operations for Tennessee Marti Miller learned of the compliance review’s findings, she took over

the investigation, overruled many of the review’s findings, and released billing holds on at least

some of the relevant patients—allowing Avalon to go back to billing for those patients’ hospice

services. (Id. ¶¶ 183, 199–200.)

C. The False Claims Act and This Litigation

“Since its enactment . . . , the False Claims Act”—commonly referred to as the “FCA”—

“has authorized both the Attorney General and private qui tam relators to recover from persons

who make false or fraudulent claims for payment to the United States.” Graham Cty. Soil & Water

Conservation Dist. v. U.S. ex rel. Wilson, 559 U.S. 280, 283 (2010). The qui tam provision creates

an incentive for private individuals who become aware of fraud on the government to bring that

fraud to light, in exchange for a share of the recovery—which, pursuant to 31 U.S.C. § 3729(a)(1),

can include both treble damages and penalties. See State Farm Fire & Cas. Co. v. U.S ex rel.

Rigsby, 137 S. Ct. 436, 440 (2016) (observing that “[t]his system is designed to benefit both the

relator and the Government.”).

A qui tam complaint under the FCA “shall be filed in camera, shall remain under seal for

at least 60 days, and shall not be served on the defendant until the court so orders.” 31 U.S.C. §

3730(b)(2). “The Government may, for good cause shown, move the court for extensions of the

time during which the complaint remains under seal . . . .” 31 U.S.C. § 3730(b)(3). Eventually,

however, the Government faces a choice: it must either inform the court that it wishes to “proceed

with the action, in which case” the Government will take over the litigation; or it must “notify the

court that it declines to take over the action, in which case the person bringing the action shall have

the right to conduct the action.” 31 U.S.C. § 3730(b)(4)(A)–(B).

On July 9, 2013, four qui tam relators—all of them former employees of Avalon hospice

agencies in Tennessee—jointly filed an FCA Complaint against Curo, Avalon, and Regency,

raising many of the issues that the court has discussed, along with a handful of other allegations

including claims for retaliatory discharge based on the former employees’ resistance to the

defendants’ practices. (Doc. No. 1 ¶¶ 9–12.) The Complaint was served on the United States, and,

on September 5, 2013, the United States filed an ex parte motion requesting a six-month extension

of the seal and intervention period, which the court granted. (Doc. Nos. 19, 21.) The United States

continued to request extensions, which the court continued to grant, for the next several years,

during which time the United States assured the court that it was diligently investigating the

allegations. Eventually, in 2017, the district judge to whom the case had been assigned resigned,

and the case was reassigned to Judge Bernard A. Friedman,3 who retained the case while the

government’s investigation continued. (See, e.g., Doc. No. 52.)

3 Judge Friedman, who sits in the Eastern District of Michigan, had volunteered to assist this district when,

for various reasons, two district judges were carrying the load for the four district judge allotment in the

district.

In the meantime, however, another, significantly overlapping qui tam case had been filed

by two other former Avalon employees on October 12, 2016—this one in the Eastern District of

Tennessee. (See Case No. 3:20-cv-168 (formerly E.D. Tenn. Case No. 4:16-cv-99), Doc. No. 1.)

In addition to allegations regarding the admission of non-terminal patients, the Eastern District

Complaint alleged, in a conclusory fashion, that the defendants “allowed [their] employees to

commit acts of hospice fraud when [their] employees paid kickbacks in the form of in-kind services

to nursing homes and other facilities.” (Id. ¶ 60.) Unlike the original case filed in this district, the

Eastern District case included, in addition to the FCA claims on behalf of the United States, claims

under the Tennessee Medicaid False Claims Act (“TMFCA”), Tenn. Code Ann. §§ 71-5-181 to -

185, in the name of the state of Tennessee. The TMFCA is a Medicaid-specific statute that largely

follows the model of the FCA, including by permitting qui tam relators to file claims as

whistleblowers and by granting the government—for these purposes, the State of Tennessee—the

opportunity to investigate the allegations while the complaint remains under seal for a statutorily

granted and then, if necessary, court-extended period of time. See Tenn. Code Ann. § 71-5-183(b).

The state and federal governments in the Eastern District case—like the federal government in the

Middle District case—sought and received extensions of the seal and intervention periods for a

number of years. (See, e.g., Case No. 3:20-cv-168, Doc. No. 40.)

On February 14, 2020, the United States, acting with the consent of the State of Tennessee

and the Eastern District relators, filed an ex parte motion requesting the transfer of that case to this

district, based, at least in part, on the fact that the earlier-filed Middle District case involved related

subject matter. (Case No. 3:20-cv-168, Doc. No. 74.) That motion was granted on February 24,

2020, and the case was assigned to this judge. On September 3, 2020, the original Middle District

case was reassigned to this judge, so that the two cases could be considered together. (See Doc.

No. 85.) The court consolidated the cases. (Doc. No. 89.)

Finally, on March 2, 2021, the governments filed separate but similar Notices, informing

the court that they had made the election to intervene in part and to decline to intervene in part in

the cases. (Doc. Nos. 91–92.) The United States “intervene[d] in that part of the action which

alleges that the defendants violated the FCA, 31 U.S.C. [§] 3729(a)(1)(A),by submitting, or by

causing the submission of claims, to Medicare and Medicaid for services provided to patients who

were not eligible for the federal hospice benefit because they were not terminally ill,” but it

declined to intervene with regard to any other allegations, including the kickback-related

allegations. (Doc. No. 91 at 2.) Tennessee’s election was essentially the same, other than being

limited to Medicaid: the state intervened as to the allegations regarding the admission of ineligible

patients but otherwise declined. (Doc. No. 92 at 2.)

On June 1, 2021, the governments jointly filed their Consolidated Complaint in

Intervention. (Doc. No. 100.) It states six causes of action, based on events from “at least January

1, 2010 through February 20, 2020.” (Doc. No. 100 ¶ 5.) The first two causes of action are stated

by the United States under the FCA: first, for presenting or causing the presentation of false claims,

pursuant to 31 U.S.C. § 3729(a)(1)(A); and, second, for improperly retaining overpayments,

pursuant to 31 U.S.C. § 3729(a)(1)(G). (Id. ¶¶ 376–381.) The next two causes of action are on

behalf of the State of Tennessee and state the same theories of liability under the TMFCA. (Id. ¶¶

382–87.) The final two causes of action are stated on behalf of both governments and rely,

respectively, on theories of payment under mistake of fact and unjust enrichment.4 (Id. ¶¶ 388–

93.)

4 Typically, a claim for unjust enrichment is based on state common law. However, courts have recognized

that a federal ‘common law’ claim for unjust enrichment is available surrounding certain federal statutes or

When the government intervenes in an FCA case, “it shall have the primary responsibility

for prosecuting the action, and shall not be bound by an act of the person bringing the action.” 31

U.S.C. § 3730(c)(1). However, “[i]f the Government elects not to proceed with the action, the

person who initiated the action shall have the right to conduct the action.” 31 U.S.C. § 3730(c)(3).

The TMFCA functions the same way. See Tenn. Code Ann. § 71-5-183(c)(2)–(3). Because the

governments only intervened in portions of the pending cases but declined to intervene in others,

the relators—whom the court will refer to as the “Middle District relators” and the “Eastern District

relators” (although all claims are now in the Middle District)—had a right to pursue the remaining

claims. Each set of relators filed Amended Complaints stating their claims as they stood post-

intervention. (Doc. Nos. 98, 102.) The Eastern District relators amended their complaint two more

times with leave of the court. (Doc. Nos. 142, 160.)

The defendants filed a Motion to Dismiss addressed to the governments’ claims.5 (Doc.

No. 136.) The motion does not ask the court to dismiss those claims in their entirety. Rather, the

defendants ask the court to dismiss “portions of” the governments’ claims in the following ways:

[1] Dismiss for failure to state a claim all claims involving the 24 Avalon locations

for which no exemplars or other sufficient allegations are provided;

[2] Dismiss for failure to state a claim all claims outside the dates ranges pleaded

for specific hospice locations, that is, all claims other than (a) Avalon Jackson (for

admissions between June 2010 and March 2012); (b) Avalon Athens (for

admissions between September 2015 and December 2016); and (c) Avalon

Tullahoma (for admissions between 2011 and 2015), unless independently

dismissed on other grounds;

payment obligations, with the federal unjust enrichment cause of action providing a backstop for liability

related to those obligations akin to the backstop that unjust enrichment and quasi-contract provide for state

common law claims that are similar to, but technically fall outside the bounds of, contract law. See, e.g.,

Farm Bureau Gen. Ins. Co. of Mich. v. Blue Cross Blue Shield of Mich., 655 F. App’x 483, 488 (6th Cir.

2016) (discussing common law unjust enrichment in the context of ERISA).

5 On November 7, 2021, the court issued an order denying the pending motions to dismiss without

prejudice to refiling, in light of amendments to some of the relators’ allegations. (Doc. No. 146.) The

defendants renewed their motion directed at the governments’ claims. (Doc. No. 147.)

[3] Dismiss for failure to state a claim all claims against Regency, Curo Holdings,

and Curo Health Services;

[4] Dismiss for failure to state a claim all claims that merely allege a disagreement

in clinical judgment, that is, all claims alleged related to Avalon Tullahoma;

[5] Dismiss for failure to state a claim Count II alleging reverse false claims; and

[6] Dismiss all claims under the Tennessee Medicaid FCA (Counts II and IV) and

for payment under mistake of fact and unjust enrichment (Counts V and VI) to the

same extent that the federal FCA claims are dismissed.

(Doc. No. 136 at 1–2.) The defendants also filed motions seeking the outright dismissal of the

relators’ claims. (Doc. Nos. 148, 165.)

II. LEGAL STANDARD

In deciding a motion to dismiss for failure to state a claim under Rule 12(b)(6), the court

will “construe the complaint in the light most favorable to the plaintiff, accept its allegations as

true, and draw all reasonable inferences in favor of the plaintiff.” Directv, Inc. v. Treesh, 487 F.3d

471, 476 (6th Cir. 2007); Inge v. Rock Fin. Corp., 281 F.3d 613, 619 (6th Cir. 2002). The Federal

Rules of Civil Procedure require only that a plaintiff provide “a short and plain statement of the

claim that will give the defendant fair notice of what the plaintiff’s claim is and the grounds upon

which it rests.” Conley v. Gibson, 355 U.S. 41, 47 (1957). The court must determine only whether

“the claimant is entitled to offer evidence to support the claims,” not whether the plaintiff can

ultimately prove the facts alleged. Swierkiewicz v. Sorema N.A., 534 U.S. 506, 511 (2002) (quoting

Scheuer v. Rhodes, 416 U.S. 232, 236 (1974)).

The complaint’s allegations, however, “must be enough to raise a right to relief above the

speculative level.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 555 (2007). To establish the

“facial plausibility” required to “unlock the doors of discovery,” the plaintiff cannot rely on “legal

conclusions” or “[t]hreadbare recitals of the elements of a cause of action,” but, instead, the

plaintiff must plead “factual content that allows the court to draw the reasonable inference that the

defendant is liable for the misconduct alleged.” Ashcroft v. Iqbal, 556 U.S. 662, 678-79 (2009).

“[O]nly a complaint that states a plausible claim for relief survives a motion to dismiss.” Id. at

679; Twombly, 550 U.S. at 556.

Rule 9(b) of the Federal Rules of Civil Procedure states that “a party must state with

particularity the circumstances constituting fraud,” if the cause of action at issue includes such an

element. The Sixth Circuit has explained that, while Rule 9(b) imposes a heightened standard, the

underlying purpose of the rule is to serve the same ends as the general pleading requirements of

Rule 8:

[Rule 9(b)] should not be read to defeat the general policy of “simplicity and

flexibility” in pleadings contemplated by the Federal Rules. Rather, Rule 9(b) exists

predominantly for the same purpose as Rule 8: to provide a defendant fair notice of

the substance of a plaintiff’s claim in order that the defendant may prepare a

responsive pleading. Rule 9(b), however, also reflects the rulemakers’ additional

understanding that, in cases involving fraud and mistake, a more specific form of

notice is necessary to permit a defendant to draft a responsive pleading

United States ex rel. SNAPP, Inc. v. Ford Motor Co., 532 F.3d 496, 504 (6th Cir. 2008) (citations

and quotation marks omitted). “So long as a [plaintiff] pleads sufficient detail—in terms of time,

place, and content, the nature of a defendant’s fraudulent scheme, and the injury resulting from the

fraud—to allow the defendant to prepare a responsive pleading, the requirements of Rule 9(b) will

generally be met.” Id. “Where a complaint alleges ‘a complex and far-reaching fraudulent scheme,’

then that scheme must be pleaded with particularity and the complaint must also ‘provide examples

of specific’ fraudulent conduct that are ‘representative samples’ of the scheme.” United States ex

rel. Marlar v. BWXT Y–12, LLC, 525 F.3d 439, 444–45 (6th Cir. 2008) (quoting United States ex

rel. Bledsoe v. Cmty. Health Sys., Inc., 501 F.3d 493, 510 (6th Cir. 2007)).

III. ANALYSIS6

A. Request to Narrow the Governments’ Claims Based on Time Period and/or Location

It is not surprising that the defendants’ motion directed at the governments’ claims is

focused on narrowing those claims, rather than asking the court to dismiss them all outright. The

plaintiffs have identified at least some actions that, if they actually occurred, are hard to

characterize as anything but fraudulent. Hospice fraud preys on medicine’s gray areas, but there is

nothing tricky or ambiguous about, for example, the leader of a facility instructing a nurse to lie

about a patient’s symptoms in order to get a claim paid. Indeed, the defendants, in their briefing,

concede that, if nothing else, the plaintiffs have adequately alleged that Avalon Jackson was, for a

time, run by an “outlier” DOO, who “resigned in the face of a compliance audit or was terminated.”

(Doc. No. 137 at 8.) The governments therefore have had little trouble pleading at least some

potentially actionable fraud claims. The questions raised by the defendants’ motion are instead

whether the government plaintiffs have pleaded their causes of action too broadly and whether the

court should, in effect, trim those causes of action down, in terms of their time periods and the

facilities involved, before this litigation goes any further.

The court will not do so. The plaintiffs’ first two proposed grounds for limiting the scope

of the governments’ claims are based on a strikingly aggressive—and unsupported—reading of

the requirement, under Rule 9(b), to file exemplary instances of fraud if a plaintiff has alleged a

wide-ranging scheme involving numerous transactions. The purpose of an example is just that: to

be an example. The defendants, however, ask the court instead to treat the governments’ examples

6 The defendants’ arguments for dismissing the governments’ unjust enrichment and payment by mistake

of fact claims are functionally identical to their arguments regarding the FCA and TMFCA. (See Doc. No.

163 at 18 (stating that the claims are “indistinguishable” and “should fail for the same reasons”).) The court,

accordingly, will direct its analysis at those statutory claims.

as, in essence, fenceposts erecting an artificial boundary outside of which the plaintiffs’ claims, no

matter how otherwise well-pleaded, can extend. Pleading examples on a few dates, the plaintiffs

argue, should mean not pursuing any fraud outside of the general range of those dates, and pleading

examples at a few facilities should mean not pursuing any fraud—even otherwise wholly identical

fraud by the same defendant—at other facilities in the same state. The defendants, however, have

not identified any Sixth Circuit caselaw suggesting that that is actually how Rule 9(b) works.

“Rule 9(b) is not designed to force a plaintiff to prove their case in the complaint . . . .”

United States v. Assocs. in Eye Care, P.S.C., No. Civ. 13-27-GFVT, 2014 WL 414231, at *7 (E.D.

Ky. Feb. 4, 2014) (citing New England Health Care Emps. Pension Fund v. Fruit of the Loom,

Inc., No. CIV.A. 1:98-CV-99-M, 1999 WL 33295037, at *6 (W.D. Ky. Aug. 16, 1999)). There is

no requirement in the text of the Rule or the caselaw applying it that a plaintiff alleging the same

pattern of fraud by the same defendants in multiple offices or facilities must plead an example for

every single one. Nor is there any requirement that a plaintiff must provide an example from on or

near the first day of the alleged scheme and another example from on or near the last day of the

alleged scheme or risk forfeiting the full relevant period of the fraud. Rule 9(b) requires a plaintiff

to plead specifically, not exhaustively.

Indeed, the Sixth Circuit has recognized that the very purpose of the “exemplary claims”

requirement is to allow a plaintiff to fully assert a scheme involving “many allegedly false claims

over a substantial period of time” without compromising the “logistical efficiency” of the pleading

process by requiring unwieldy descriptions of claim after claim after claim. Bledsoe, 501 F.3d at

509. “[A]bsent authority from this Court, the Sixth Circuit, or the Supreme Court,” there is no

basis for “hold[ing] that the failure to allege examples over the entire span of an alleged fraudulent

scheme requires dismissal of” claims supported by “otherwise representative examples that have

been plead with sufficient particularity.” U.S. ex rel. Lovett v. Holzer Clinic., No. 2:08-CV-312,

2014 WL 12767601, at *10 (S.D. Ohio Mar. 26, 2014).

It is, moreover, difficult to see where the defendants’ logic in support of their motion would

end. If the governments had pleaded examples from each hospice agency, the defendants could

object that there was not an example for each employee or each diagnosis. If the governments had

pleaded examples from every relevant year, the plaintiffs could object that there were not examples

from each month or each week. An enterprising defendant can find lines like these to draw

anywhere, no matter how well-pleaded a complaint is.

The infeasibility of such a standard becomes clear if one attempts to apply it to any number

of ordinary types of FCA case. Imagine, for example, that the United States or a relator was

pursuing a company-wide FCA case against a large national pharmacy chain. See, e.g., U.S. ex rel.

Kester v. Novartis Pharms. Corp. et al., 43 F. Supp. 3d 332, 342 (S.D.N.Y. 2014). Would Rule

9(b) require the complaint to include an example from every pharmacy location for every segment

of the underlying time period, even if that meant thousands of examples and a complaint the length

of an encyclopedia volume? The defendants have found no court that has endorsed such a reading.

Aside from the fact that so stringent a standard would serve very little purpose, it would also be

impossible to reconcile with the “general policy of ‘simplicity and flexibility’ in pleadings” that,

the Sixth Circuit has held, Rule 9(b) does not negate. SNAPP, Inc., 532 F.3d at 504.

The reason that courts allow FCA plaintiffs to satisfy Rule 9(b) with examples, rather than

a comprehensive list of individual claims, is that achieving that level of specificity and granularity

for every false claim would, at least in a case like this one, “be extremely ungainly, if not

impossible.” Bledsoe, 501 F.3d at 509 (citing U.S. ex rel. Franklin v. Parke–Davis, Div. of

Warner–Lambert Co., 147 F.Supp.2d 39, 49 (D. Mass. 2001)). A policy of requiring separate

examples for every conceivable variation on the ancillary details of the underlying claims would

pose the same problem. There is, of course, a limit to how much a plaintiff can rely on one example

to support an array of attenuated theories of liability. What qualifies as a sufficient set of examples

for any given case, however, depends on the nature of the allegations at issue, not on a rote

checklist of dates and service locations. For example, in United States ex rel. Hinkle v. Caris

Healthcare, L.P., No. 3:14-CV-212-TAV-HBG, 2017 WL 3670652 (E.D. Tenn. May 30, 2017),

the District Court for the Eastern District of Tennessee concluded that the government’s

description of “six ‘audit sample patients’ in detail” amounted to “sufficient illustrative examples

of patients whom defendants allegedly knew did not qualify for hospice treatment” for the

purposes of the twenty-six facilities at issue, because those examples, combined with the

government’s pleading of the general scheme itself, “g[ave] defendants fair notice of the FCA

claims brought against them under Rules 8 and 9(b).” Id. at *3, *9–10. The pattern in this case is

almost identical: the governments provided multiple examples to demonstrate a statewide problem

of improper hospice certification. In this case, like in that one, the examples are adequate.

The governments in this case have alleged a coherent, specific pattern whereby (1)

corporate pressure placed a thumb on the scales in favor of improper admissions and (2) the

personnel in individual agencies predictably responded by admitting and/or retaining patients who

did not qualify for hospice by any reasonable clinical standard. Although some Avalon agencies

may have been worse than others, the top-down, corporate policies that drove all of the alleged

false claims were not limited to any specific Avalon location, and the plaintiffs have provided

examples from multiple Avalon locations, at multiple times, which illustrate the ways in which the

scheme resulted in fraud against the Medicare and Medicaid systems. In the absence of some

specific, binding authority suggesting that such pleading was nevertheless insufficient, the court

will not prematurely dismiss any aspect of the plaintiffs’ claims based on supposed noncompliance

with pleading requirements.7

B. Claims Involving Disagreements of Clinical Judgment

The defendants argue next that the court should dismiss the governments’ “claim[s] for

FCA violations based solely on disagreements with certifying physician’s clinical judgments.”

(Doc. No. 137 at 16.) The defendants are probably correct that, if the governments or relators had

actually sought to assert any such claims—that is, claims alleging that it is fraudulent merely to

make a good-faith clinical determination of terminal illness with which the government

disagrees—those claims would be appropriate for dismissal. “[A] reasonable difference of opinion

among physicians reviewing medical documentation ex post is not sufficient on its own to suggest

that those judgments—or any claims based on them—are false under the FCA.” United States v.

AseraCare, Inc., 938 F.3d 1278, 1297 (11th Cir. 2019). The governments, however, do not allege

a mere clinical disagreement; they allege fraud, in the simplest and most straightforward sense.

Admittedly, that fraud involved prognosis determinations that required the exercise of judgment,

but it is well-settled that “opinions are not, and have never been, completely insulated from

scrutiny” for fraud. United States v. Paulus, 894 F.3d 267, 275 (6th Cir. 2018) “At the very least,

opinions may trigger liability for fraud when they are not honestly held by their maker, or when

the speaker knows of facts that are fundamentally incompatible with his opinion.” Id. (citations

omitted). That is what the governments allege happened at Avalon.

7 From the briefing, it appears that the defendants’ request to narrow the scope of the case is motivated, at

least in part, by a desire to limit excessively burdensome or intrusive discovery. The court notes that this

ruling does not prevent them from objecting to the breadth or burden of any specific discovery request. Nor

does the court’s Rule 12(b)(6) analysis prevent the defendants, when making any such objection, from

noting that the necessity of any government discovery request should be assessed in light of the fact that

the United States has already had nearly nine years to investigate these allegations unilaterally.

The defendants direct this argument at Patient No. 4 in particular, on the assumption that,

if the billing for Patient No. 4 cannot actually support an FCA claim, then the court must dismiss

all FCA claims arising out of Avalon Tullahoma. As the court has already held, the defendants’

attempts to impose a rigid location-by-location pleading requirement on the governments’ claims

is unsupported and unpersuasive. In any event, though, it is not true that the governments’ pleading

with regard to Patient No. 4 alleges only a difference in clinical judgment. To the contrary, the

governments specifically allege that Patient No. 4’s “medical records do not support a diagnosis

of end-stage liver failure or any other terminal diagnosis” at any point during the patient’s time in

Avalon’s care. (Doc. No. 100 ¶ 303.) The governments further allege that, “upon admission and at

multiple points during Patient No. 4’s stay, the patient’s lab results failed to meet clinical indicators

for end-stage liver disease, such as those set forth in the relevant LCD.” (Id.) They allege that

Avalon’s own staff eventually “decided to discharge the patient because he “appear[ed] more

chronic than terminal.” (Id. ¶ 321.) Moreover, while a patient’s exceeding his predicted six-month

prognosis is not inherently enough to render a hospice claim fraudulent, it can be corroborative,

and the governments have alleged that Patient No. 4’s liver disease did not, in fact, turn out to be

in its end stage—if “end stage” means six months—because he ended up surviving for over five

more years. (Id. ¶ 323.) Without the court’s reiterating every detail, it suffices to say that the

governments have pleaded that Patient No. 4 was not terminally ill, as defined under Medicare or

TennCare, and that the defendants knowingly or recklessly certified him as such and billed those

programs for his hospice services anyway. That is precisely what the governments needed to plead

to survive a Rule 12(b)(6) motion.

It is entirely possible that, when the parties have the opportunity to present evidence, the

defendants will be able to demonstrate that the certification of Patient No. 4’s terminal illness was

reasonable, or at least sufficiently within the bounds of ordinary clinical judgment that the

certification was not recklessly false. Not every good argument, however, is a good argument

under Rule 12(b)(6). Because the defendants are asking the court to depart from its duty, at this

stage, merely to test the pleadings, their request will be denied.

C. Reverse False Claims

The defendants argue next that the court should dismiss the governments’ claims based on

the so-called “reverse false claims” provisions of the FCA and TMFCA. See 31 U.S.C. §

3729(a)(1)(G), Tenn. Code Ann. § 71-5-182(a)(1)(D). The FCA’s reverse false claims provision,

generally speaking, prohibits the knowing avoidance of an obligation to pay money to the United

States. In a case involving a government healthcare program, that provision, codified at 31 U.S.C.

§ 3729(a)(1)(G), “requires [the plaintiff] to allege facts that show defendants,” acting with the

appropriate culpable mental state, “received overpayments from the government and failed to

refund those payments.” U.S. ex rel. Ibanez v. Bristol-Myers Squibb Co., 874 F.3d 905, 916 (6th

Cir. 2017) (citing 31 U.S.C. § 3729(a)(1)(G); U.S. ex rel. Prather v. Brookdale Senior Living

Communities, Inc., 838 F.3d 750, 774 (6th Cir. 2016)); see also 31 U.S.C. § 3729(b)(3) (defining

“obligation,” to include any “established duty, whether or not fixed, arising from . . . . the retention

of any overpayment”). In the alternative, liability under that section can be supported by “‘proof

that the defendant made a false record or statement at a time that the defendant owed to the

government an obligation’ . . . to pay money or property.” Id. (quoting Chesbrough v. VPA, P.C.,

655 F.3d 461, 473 (6th Cir. 2011)). The TMFCA includes a provision imposing liability on the

same two alternative grounds. Tenn. Code Ann. § 71-5-182(a)(1)(D). The governments rely

primarily on the first option, retention of overpayments, although the two grounds can overlap

significantly.8

The term “obligation,” as it is used in the FCA’s reverse false claims provision, expressly

includes any “established duty, whether or not fixed, arising from . . . . the retention of any

overpayment. 31 U.S.C.A. § 3729(b)(3). By statute, Medicare requires that “[a]n overpayment

must be reported and returned” to the program no later than “the date which is 60 days after the

date on which the overpayment was identified.” 42 U.S.C. § 1320a-7k(d)(2). Similarly, TennCare

Policy PI 11-001, which governs “Reporting and Returning Overpayments,” states that an

overpayment from TennCare must be reported and returned within 60 days from when it is

identified. (Doc. No. 100 ¶ 32.9)

CMS has provided the following guidance regarding what it means to have “identified” an

overpayment:

A person has identified an overpayment when the person has, or should have

through the exercise of reasonable diligence, determined that the person has

received an overpayment and quantified the amount of the overpayment. A person

should have determined that the person received an overpayment and quantified the

amount of the overpayment if the person fails to exercise reasonable diligence and

the person in fact received an overpayment.

42 C.F.R. § 401.305(a)(2). The defendants suggest that CMS’s definition is likely to be set aside

as inconsistent with the statutory text, because the D.C. Circuit affirmed a district court’s

8 The issue of reverse false claims liability may be ultimately redundant in many instances—including many

in this case—because the initial claims for payment will satisfy all the requirements of FCA or TMFCA

liability. It is, however, possible to be liable for a reverse false claims violation without having committed

a prior initial violation—most prominently, in a case in which an overpayment was not originally made due

to fraud but was fraudulently retained after it was identified. It is therefore permissible, and often

appropriate, to plead reverse false claims liability as an alternative theory to recovery. See United States v.

Omnicare, Inc., No. 1:15-CV- 4179 (CM), 2021 WL 1063784, at *12 (S.D.N.Y. Mar. 19, 2021) (noting

that reverse false claims “can serve as an alternative and independent basis for liability”).

9 Policy available at https://www.tn.gov/content/dam/tn/tenncare/documents2/pi11001.pdf.

conclusion to that effect with regard to a similar CMS definition in another regulation, 42 C.F.R.

§ 422.326(c), in UnitedHealthcare Ins. Co. v. Becerra, 16 F.4th 867, 893 (D.C. Cir. 2021). The

defendants argue that the court should instead construe the definition of “identified” to require

actual knowledge or, at the very least, recklessness. (See Doc. No. 163 at 15–16.)

Relying on that narrower definition, the defendants argue that the court should dismiss all

of the governments’ reverse false claims counts because the Complaint in Intervention “points to

no identified overpayments.” (Doc. No. 137 at 21.) The plaintiffs, however, have explicitly

pleaded, not only that numerous patients were improperly admitted into Avalon’s hospice care,

but that auditing and oversight imposed by the defendants at the corporate level actually uncovered

many such instances but that the defendants chose not to return the overpayments. The details

regarding retained overpayments to Avalon Jackson are particularly extensive and include the

allegation that “[a] chart audit of approximately seventeen patients revealed that at least six patients

were not hospice-appropriate,” but the company’s Vice President of Clinical Operations

“overruled the findings of Curo’s compliance staff.” (Doc. No. 100 ¶¶ 196, 200.) The governments

also detailed how Curo/Regency/Avalon personnel, at various times, noticed facts suggesting that

Patients Nos. 1 through 5 were not terminal, but the money paid for those patients was not returned,

even after some had been released from hospice altogether. Accordingly, even if the court were to

ignore the still-in-effect CMS definition and adopt the defendants’ proposed narrow reading of

“identified”—indeed, even if the court were to adopt the strictest reading possible, requiring

nothing short of actual knowledge—the governments’ pleading would still be sufficient.10

10 Because the viability of the CMS definition is not determinative, the court does not resolve this issue.

D. Claims Against Regency and Curo

The defendants argue next that, even if the plaintiffs have successfully pleaded causes of

action against Avalon itself, the court should dismiss the claims against Regency and Curo, as its

parent companies, because they were not actually the entities evaluating individual patients, filling

out COTIs, or filing claims for payments. The governments respond that they have pleaded, in

significant detail, how Regency’s, and later Curo’s, actions directly contributed to the submission

of false claims and that the court, therefore, has no basis for dismissing those claims.

The provisions of the FCA at issue in this case “attach[] liability, not to the underlying

fraudulent activity . . . , but to the claim for payment” or, in the case of reverse false claims, the

retention of funds. Sanderson v. HCA-The Healthcare Co., 447 F.3d 873, 877–78 (6th Cir. 2006)

(quoting United States v. Rivera, 55 F.3d 703, 709 (1st Cir. 1995)). The defendants are therefore

correct that, even if unlawful activities took place and even if those activities benefited a particular

defendant, the FCA will not provide the appropriate enforcement mechanism unless the defendant

has been sufficiently tied to a claim for payment under, or retention of funds owed to, a government

program. As the court has already discussed, however, the FCA provides multiple avenues through

which the connection between the defendant and the claim can be made, including by showing that

the defendant caused another party to submit a false claim. The FCA reaches not only the entity

or person who submits a false claim, but one who “causes” a false or fraudulent claim “to be

presented” or “causes to be made or used . . . a false record or statement material to an obligation

to pay or transmit” an overpayment. 31 U.S.C. § 3729(a)(1)(A), (G). That is precisely what the

governments have alleged—that Curo and Regency caused Avalon’s false claims.

The defendants do not dispute that the FCA imposes causation-based liability on non-

submitters. They argue, however, that the court should nevertheless dismiss these claims, because

the parent companies’ corporate practices, such as financial incentives and company-wide

documentation policies, were not inherently unlawful. But the governments’ claims are not

premised on any assumption that such policies are, in general, automatically unlawful. Rather, the

governments argue that the parent companies’ policies, in these particular instances, involving this

particular type of claim and services, and in light of the information contemporaneously available

to Regency and Curo, resulted in Regency’s and Curo’s knowingly or recklessly causing the

submission of specific claims that were false. There is simply no requirement, under the FCA, that

a defendant’s culpable actions be independently illegal for some reason other than those set out in

the FCA itself.

As with the issue of clinical judgment, the defendants will have every opportunity to

establish that their corporate policies were above-board and that any false claims that were

submitted by Avalon were purely Avalon’s fault. The defendants’ eagerness to make such an

argument, moreover, is understandable, given how long they have been under investigation

without the opportunity for formal rebuttal. From the perspective of the Federal Rules, however,

this case is still at the pleading stage, and the court must treat it as it would any other fraud case

challenged pursuant to Rule 12(b)(6). Because the governments have clearly and with particularity

pleaded the elements necessary to establish liability on behalf of Avalon’s corporate parents, the

court cannot dismiss those claims.

E. Claims of the Middle District Relators

1. Claims in Which the United States Intervened. The Middle District relators’

Amended Complaint, in addition to pleading claims for retaliation and for some types of FCA

liability that were not subject to government intervention, reiterates the core allegations regarding

non-terminal patients that were the subject of the United States’ intervention. The defendants argue

that those claims “should be dismissed or stayed” in light of the intervention. (Doc. No. 139 at 3.)

The defendants are correct that FCA claims do not belong to the relator. Rather, the FCA

“effect[s] a partial assignment of the Government’s damages claim” to the relator, who merely

serves as the claim’s steward unless and until the government intervenes. Vt. Agency of Nat. Res.

v. U.S. ex rel. Stevens, 529 U.S. 765, 773 (2000). At the same time, however, the FCA does not

consider the relator to be persona non grata merely because the government has taken control of

the claim. To the contrary, the relator has a “right to continue as a party to the action,” including

the rights to receive notice of certain events, to object to the government’s decision to abandon or

settle the case, and to participate in aspects of the litigation unless the court restricts that

participation. 31 U.S.C. § 3730(c)(2).

When the relators pleaded their FCA claims, they did so on behalf of the government.

Those claims, in other words, were already the government’s claims—that is, the claims that are

now proceeding forward, including those under the government’s control. All that has happened

is that the “primary responsibility for prosecuting the action” with regard to some claims has

formally vested in the government. Id. No new claims were created, and no old claims were left

behind. This aspect of the defendants’ request, therefore, is unnecessary and will be denied.

2. “False Records” Claims. The FCA includes two provisions premising liability on the

creation or use of a “false record or statement,” distinct from traditional liability based on the

presentment of a false claim for payment. The first provision, 31 U.S.C. § 3729(a)(1)(B), imposes

liability on any person or entity who “knowingly makes, uses, or causes to be made or used, a false

record or statement material to a false or fraudulent claim.” The second provision, which the court

has already discussed, is a portion of the reverse false claims provision, 31 U.S.C. § 3729(a)(1)(G).

That prohibition reaches anyone who “knowingly makes, uses, or causes to be made or used, a

false record or statement material to an obligation to pay or transmit money or property to the

Government.” Id. Although the United States intervened with regard to the Middle District

relators’ reverse false claims allegation, the United States did not intervene with regard to 31

U.S.C. § 3729(a)(1)(B), despite that provision having been cited in the Middle District relators’

initial Complaint. (Doc. No. 1 ¶ 70.) Those relators therefore could pursue a claim pursuant to 31

U.S.C. § 3729(a)(1)(B)—and they, in fact, initially indicated that they intended to do so.

However, after the defendants moved for dismissal of any such claims, the relators “elected

not to pursue a separate, free-standing false records claim under 31 U.S.C. § 3729(a)(1)(B).” (Doc.

No. 154 at 2.) With the consent of the United States, they voluntarily dismissed those claims on

November 23, 2021. (Doc. No. 167.) This portion of the defendants’ motion is therefore moot, as

the claims against which it is directed are no longer pending.

3. FCA Retaliation Claims. The FCA contains an anti-retaliation provision, which

provides a cause of action for damages and other relief to any employee who is discharged or

otherwise discriminated against “because of lawful acts done by the employee . . . in furtherance

of an action under this section or other efforts to stop [one] or more violations of” the Act. 31

U.S.C. § 3730(h)(1). Retaliatory discharge claims under the FCA proceed under the same rules

applicable to other employment-related retaliation claims, which means that a plaintiff “must

show: (1) he engaged in a protected activity; (2) his employer knew that he engaged in the

protected activity; and (3) his employer discharged or otherwise discriminated against the

employee as a result of the protected activity.” Yuhasz v. Brush Wellman, Inc., 341 F.3d 559, 566

(6th Cir. 2003) (citing McKenzie v. BellSouth Telecomms., Inc., 219 F.3d 508, 513–14 (6th Cir.

2000)).

Because the FCA’s anti-retaliation provision is specific to that statute, the underlying

protected activity must be “in furtherance” of either an FCA action itself or “other efforts” directed

at stopping an FCA violation—that is, stopping one of the various alternative types of fraud against

the government outlawed by the Act. 31 U.S.C. § 3730(h)(1). That latter route to liability—

establishing that the defendant took action against the employee based on his or her anti-fraud

efforts other than merely participating in an FCA case—was added to the statute in 2009. See

Fraud Enforcement and Recovery Act of 2009 (FERA), PL 111–21, May 20, 2009, 123 Stat 1617.

“As a result of this amendment, the FCA ‘now protects two categories of conduct.’” Mikhaeil v.

Walgreens Inc., No. 13-14107, 2015 WL 778179, at *7 (E.D. Mich. Feb. 24, 2015) (quoting Tibor

v. Mich. Orthopaedic Inst., 72 F. Supp. 3d 750, 761 (E.D. Mich. 2014)).

The defendants concede that “the pleading standards for an FCA retaliation claim are not

subject to Rule 9(b),” but they argue that the Middle District relators failed to adequately plead

that their dismissals were caused by their participation in protected activity, even pursuant to the

more forgiving pleading standards of Rule 8 alone. (Doc. No. 139 at 12.) As the plaintiffs point

out, however, aspects of the defendants’ argument rely on holdings that have, on their face, been

superseded by the 2009 amendment. (See id. at 10 (citing McKenzie, 219 F.3d at 516, for

proposition that “protected activity” must be “activity which reasonably could lead to a viable FCA

action”). Nevertheless, even under the amended, more expansive version of the provision, the

plaintiffs were required to state facts sufficient to support their claims. The court, accordingly, will

consider the allegations regarding each Middle District relator individually.

Teague alleges that she “was terminated by Defendants for refusing to perform the illegal

tasks Defendants required her to do and for reporting Defendants’ illegal activities.” (Doc. No. 98

¶ 41.) Specifically, “[s]oon after Teague made numerous complaints to her superiors making clear

references to activities occurring in the Jackson office that she was ethically and legally prohibited

from doing, Defendants terminated her employment” for a reason that, she says, was pretextual.

(Id.) The Complaint lists multiple specific illegal activities in which Teague refused to participate

and about which she alerted her superiors, including several that could form the basis of an FCA

claim. (Id. ¶ 42.) These allegations plainly satisfy the requirements of alleging retaliatory

discharge.

Pence similarly specifically alleges that she “was terminated . . . for refusing to perform

the illegal tasks Defendants required her to do and for reporting Defendants’ illegal activity.” (Id.

¶ 47.) The Complaint specifically lists the fraudulent activities involved, as well as the process by

which, according to Pence, the defendants built a pretextual case for her termination. (Id. ¶ 49.)

She also cites comments from “corporate officials” during that process that appear to have been

veiled criticisms of her whistleblowing, packaged as complaints about her negativity,

insubordination, or dishonesty. (Id. ¶ 49.) These allegations, like Teague’s, plainly satisfy the

requirements of alleging protected activity and causation.

Adkins’ allegations are even more plainly sufficient. She states that she “was placed on

administrative leave after she reported illegal activities to her superiors, refused to participate in

the illegal activity, and then reported to the Tennessee Bureau of Investigation about Defendants’

illegal activity.” (Id. ¶ 53.) She states that she was eventually denied a promotion for which she

was qualified and eventually terminated based on her actions. (Id. ¶¶ 56–58.) She even quotes a

supervisor saying to her, “we don’t want any whistleblowers.” (Id. ¶ 56.) Such allegations are more

than sufficient to survive a Rule 12(b)(6) motion.

Finally, Carnell alleges that she was constructively terminated after having “made

numerous complaints to her superiors making clear references to activities occurring in the Jackson

office that were fraudulent and illegal and in which she was ethically and legally prohibited from

participating.” (Id. ¶ 60.) Like the other Middle District relators, she lists the unlawful practices

involved, at least some of which could form the predicate for FCA claims. (Id. ¶ 61.) She explains

that her constructive termination occurred by her having to choose between complying with her

ethical obligations as a licensed nurse and following directions from Avalon. (Id. ¶ 62.) Her FCA

retaliation claim was therefore sufficiently pleaded and will not be dismissed.

4. State Law Retaliation Claims. The Middle District relators have “elected not to pursue

[their] redundant state law claims,” which the defendants have argued are untimely (Doc. No. 154

at 6), and they voluntarily dismissed those claims along with their “false records” claims. (Doc.

No. 167.) This portion of the defendants’ motion is therefore similarly moot.

5. Claims Against Regency and Curo. Finally, the defendants argue that, even if the

Middle District relators’ claims against Avalon should be permitted to proceed, the court should

dismiss the claims against Regency and Curo. The relators, however, have alleged that all of the

corporate defendants were alter egos for the purposes of FCA retaliation and pleaded specific facts

in support of that contention. (Doc. No. 98 ¶¶ 15–16.) Whether they will prevail on such an

argument is an open question. At this stage, however, there is no basis for prematurely dismissing

the claims against the other corporate entities for their alleged involvement in the retaliation,

particularly given that such claims are not subject to Rule 9(b) and therefore may be pleaded more

generally.

F. Claims of the Eastern District Relators

1. Kickback-Based FCA Claims. The federal Anti-Kickback Statute (“AKS”) makes it

unlawful to

knowingly and willfully offer[] or pay[] any remuneration (including any kickback,

bribe, or rebate) directly or indirectly, overtly or covertly, in cash or in kind to any

person to induce such person . . . to refer an individual to a person for the furnishing

or arranging for the furnishing of any item or service for which payment may be

made in whole or in part under a Federal health care program, or . . . to purchase,

lease, order, or arrange for or recommend purchasing, leasing, or ordering any

good, facility, service, or item for which payment may be made in whole or in part

under a Federal health care program . . . .

42 U.S.C. § 1320a-7b(b)(2).”In short, a kickback violation entails 1) remuneration to a person or

entity in a position to refer Federal health care program patients 2) that could reasonably induce

the person or entity to refer such patients.” Jones-McNamara v. Holzer Health Sys., 630 F. App’x

394, 401 (6th Cir. 2015) (citing OIG Supplemental Compliance Program Guidance for Hospitals,

70 Fed. Reg. at 4864). “Federal health care program” includes Medicare and Medicaid. 42 U.S.C.

§ 1320a-7b(f).

Although the AKS “is a criminal statute,” AKS violations frequently form the basis for

civil claims under the FCA. U.S. ex rel. Arnstein v. Teva Pharm. USA, Inc., No. 13 CIV. 3702

(CM), 2019 WL 1245656, at *5 (S.D.N.Y. Feb. 27, 2019) (citing Donovan v. Rothman, 106 F.

Supp. 2d 513, 516 (S.D.N.Y. 2000)). The linkage between the two statutes was first pursued by

governments and relators themselves in the context of litigation, but it was formalized by statutory

amendment in 2010. Id. Pursuant to that amendment, a claim to Medicare or Medicaid “that

includes items or services resulting from a violation of [the AKS] constitutes a false or fraudulent

claim” as a matter of law. 42 U.S.C. § 1320a-7b(g).

The Eastern District relators allege that the defendants’ marketers “promised the owners of

[two assisted living] facilities the use of CNAs (certified nursing assistants) and offered [various]

types of supplies if the owner would agree for the residents to receive hospice services,” and that

“Avalon would provide CNAs to assist with daily care, and supplies including, but not limited to,

bed pads, diapers, skin barrier cream, shampoos, toothpaste, and medication and other items,” in

violation of the AKS. (Doc. No. 160 ¶ 58.) The Eastern District relators argue that such practices

violated the AKS and gave rise to FCA claims. The governments declined to intervene in any such

claims, meaning that the Eastern District relators are solely responsible for pleading them

sufficiently.

An AKS-based FCA claim is still an FCA claim, meaning that it sounds in fraud and must

be pleaded in accordance with Rule 9(b), including by pleading at least one exemplary claim. The

Eastern District Relators do identify a number of patients who, they claim, were improperly

admitted to Avalon hospice. (Doc. No. 160 ¶¶ 36–56.) None of those examples, however, includes

an express allegation that the patients were referred pursuant to a kickback relationship. A few of

the patients—namely, Patients 28, 31, and 35—are expressly identified as having been patients of

Extended Family Care, one of the two facilities with which Avalon allegedly had an improper

kickback relationship. (Id. ¶¶ 52–53, 55.) That comes closer to satisfying Rule 9(b), but the

defendants do not allege that any of those patients was actually referred by Extended Family Care

or that the underlying claims would not have occurred but for kickbacks. (Id.) The FCA only

reaches claims “that include[] items or services resulting from” an underlying AKS violation. 42

U.S.C. § 1320a-7b(g) (emphasis added). It is not enough merely to have alleged that Patients 28,

31, and 35 happened to have received care from two entities with an improper relationship. The

claims to government healthcare programs for those patients must themselves have been tainted

by that relationship. The Eastern District relators have not alleged that they were.

This analysis may seem like it is splitting hairs or asking too much of the relators. But these

relators are already on their Third Amended Complaint. When the court granted them leave to file

that most recent version of their allegations, the court expressly stated that it did so “reluctantly”

and would “not permit these relators another opportunity to amend, absent extraordinary

circumstances warranting such a step.” (Doc. No. 159.) It is not expecting too much to ask a party

to specifically state the basic elements of a claim on the party’s fourth try. To the contrary, it is

what Rule 9(b) requires. The court, accordingly, will dismiss the FCA claims based on the AKS.

2. Retaliation Claims. The Eastern District relators, like the Middle District relators, have

pleaded claims under the FCA’s retaliation provision. The same standards that the court has

already discussed govern those claims. The court will apply those standards to the allegations of

Anderson and Mathis individually, as it did with the other relators.

Anderson alleges that she “expressed opposition in the IDG meeting to the Defendant’s

medical director, Dr. Ephraim Gammada, Nurse Practitioner Deanna Britton and Director of

Operations Nena Hart that an Alzheimer’s diagnosis was not appropriate for hospice care for a

patient who had a good memory, remained alert, oriented and was ambulatory,” but that “[t]he

medical director took no action to change the certificate of illness and allowed the patient to remain

on hospice services.” (Doc. No. 160 ¶ 80.) She describes raising specific concerns, on more than

one occasion, about Alzheimer’s patients’ having been inappropriately diagnosed as having

reached a terminal stage. In one such instance, the patient’s diagnosis was simply changed to

suffering from a terminal stage of a different progressive illness, COPD. (Id. ¶¶ 77–88.)

On the same day as that last instance, Anderson was terminated for allegations of

misconduct that, she says, were false. (Id. ¶¶ 89–90.) Anderson expressly alleges that the “reasons

given to terminate [her] were a mere pretext to discharge her because of her opposition and/or

refusal to participate in, or for refusing to remain silent about illegal activities involving fraudulent

practices and submission of false claims.” (Id. ¶ 94.) While Anderson perhaps could have done

more to be as explicit as possible to link her termination to seeking to prevent false claims—as

opposed to seeking to prevent general wrongdoing—she is entitled to all reasonable inferences in

her favor, and the allegations involve issues of terminal illness certification that, both under the

law and in the context of the pleading, provide a plain nexus with the FCA. Anderson’s, therefore,

allegations state a claim for retaliation under the FCA.

Mathis alleges that, “[i]n an effort [to] stop violations of the False Claims Act, [he]

expressed his concern and opposition about promises of kickbacks and false certifications with

Avalon/Curo’s Director(s) of Operations and Area Director of Operations” and “met with two

nursing home administrators concerning complaints about Avalon/Curo’s marketers making

referrals for patients who clearly were not qualified for hospice care.” (Id. ¶ 60.) Shortly thereafter,

Mathis was fired, allegedly for various types of misconduct and substandard performance. (Id. ¶¶

70–73.) He alleges that these reasons were pretextual, reflected a combination of falsehoods and

an inconsistent application of the agency’s employee standards, and concealed the fact that he was

terminated for speaking up about the agency’s fraud. (Id. ¶ 71–74.) These allegations state a claim

for unlawful retaliatory discharge under the FCA.

The defendants point out that, according to Anderson’s and Mathis’s allegations, there

were ostensible reasons for firing them other than their fraud-related actions. That, though, is true

in virtually every wrongful termination case. While the defendants will have every opportunity to

argue that their reasons were not pretextual, that inquiry is beyond the scope of a Rule 12(b)(6)

motion. The court, accordingly, will not dismiss these relators’ claims for retaliation under the

FCA.

IV. CONCLUSION

For the foregoing reasons, the defendants’ Renewed Motion to Dismiss the Consolidated

Complaint in Intervention of the United States and Tennessee (Doc. No. 147) will be denied, their

Renewed Motion to Dismiss the First Amended Complaint of Plaintiff-Relators Teague, Pence,

Adkins, and Carnell (Doc. No. 148) will be denied in part on the merits and in part as moot, and

their Motion to Dismiss the Third Amended Complaint of Plaintiff-Relators Rachel Carol

Anderson and Steven Troy Mathis (Doc. No. 165) will be granted in part and denied in part. The

FCA claims based on violations of the AKS will be dismissed.

An appropriate order will enter.

ALETA A. TRAUGER

United States District Judge

40