DaVita, Inc. v. Marietta Mem. Hosp.

978 F.3d 326

• Court: Court of Appeals for the Sixth Circuit
• Decided: October 14, 2020
• Docket: 19-4039
• Status: Published

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 20a0328p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

DAVITA, INC.; DVA RENAL HEALTH, INC., ┐ Plaintiffs-Appellants, │ │ > No. 19-4039 v. │ │ │ MARIETTA MEMORIAL HOSPITAL EMPLOYEE HEALTH │ BENEFIT PLAN; MARIETTA MEMORIAL HOSPITAL; │ MEDICAL BENEFITS MUTUAL LIFE INSURANCE CO., │ Defendants-Appellees. │ ┘

Appeal from the United States District Court for the Southern District of Ohio at Columbus. No. 2:18-cv-01739—Sarah Daggett Morrison, District Judge.

Argued: July 30, 2020

Decided and Filed: October 14, 2020

Before: MOORE, CLAY, and MURPHY, Circuit Judges.

_________________

COUNSEL

ARGUED: Bobby R. Burchfield, KING & SPALDING LLP, Washington, D.C., for Appellants. William H. Prophater, Jr., NEWHOUSE, PROPHATER, KOLMAN & HOGAN, LLC, Columbus, Ohio, for Marietta Memorial Appellees. Rodney A. Holaday, VORYS, SATER, SEYMOUR AND PEASE LLP, Columbus, Ohio, for Appellee Medical Benefits Mutual Life Insurance Co. ON BRIEF: Bobby R. Burchfield, KING & SPALDING LLP, Washington, D.C., for Appellants. William H. Prophater, Jr., D. Wesley Newhouse, NEWHOUSE, PROPHATER, KOLMAN & HOGAN, LLC, Columbus, Ohio, for Marietta Memorial Appellees. Rodney A. Holaday, VORYS, SATER, SEYMOUR AND PEASE LLP, Columbus, Ohio, Brent D. Craft, VORYS, SATER, SEYMOUR AND PEASE LLP, Cincinnati, Ohio, for Appellee Medical Mutual Benefits Mutual Life Insurance Co. Deanna J. Reichel, Ryan V. Petty, FISH & RICHARDSON P.C., Minneapolis, Minnesota, Mary L. Stoll, STOLL LAW GROUP, No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 2

PLLC, Seattle, Washington, John R. Christiansen, CHRISTIANSEN IT LAW, Olympia, Washington, for Amici Curiae.

MOORE, J. (pp. 2–33; app. 34–38), delivered the opinion of the court in which CLAY, J., joined. MURPHY, J. (pp. 39–55), delivered a separate opinion concurring in the judgment in part and dissenting in part. _________________

OPINION _________________

KAREN NELSON MOORE, Circuit Judge. Plaintiff DaVita, Inc. and its subsidiary, DVA Renal Healthcare, Inc., appeal the district court’s dismissal of their lawsuit alleging various violations of the Medicare Secondary Payer Act and the Employee Retirement Income Security Act of 1974 by an employee health benefit plan and its administrators. According to DaVita, the plan unlawfully treated a plan participant and DaVita patient—known as Patient A in this lawsuit—differently because this patient suffers from end-stage renal disease. In particular, the plan allegedly targeted renal dialysis services, which DaVita provides to Patient A, with poor reimbursement rates, in the hopes that dialysis patients like Patient A would switch to Medicare, which they are legally entitled to do three months after being diagnosed with the disease. Upon the defendants’ motions to dismiss, the district court dismissed all of DaVita’s claims with prejudice, and DaVita appealed. For the following reasons, we are persuaded that, as to Counts I, II, and VII of its complaint, DaVita has plausibly alleged that the defendants have engaged in unlawful discrimination. As to the rest of its claims, DaVita lacks a sufficient interest to prosecute them. Accordingly, we AFFIRM in part, REVERSE in part, and REMAND for discovery and further proceedings on Counts I, II, and VII of DaVita’s complaint.

I. BACKGROUND1

Plaintiff DaVita, and its subsidiary, Plaintiff DVA Renal Healthcare, Inc., are leading providers of dialysis treatment in the United States. R. 1 (Compl. ¶¶ 11–12) (Page ID #5). Since April 15, 2017, DaVita has provided dialysis treatment to Patient A, an anonymous individual

1The following facts are undisputed unless otherwise noted. All facts are construed in the light most favorable to the plaintiff. See Lindenberg v. Jackson Nat’l Life Ins. Co., 912 F.3d 348, 357 (6th Cir. 2018). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 3

diagnosed with end-stage renal disease (“ESRD”). Id. ¶¶ 19, 29 (Page ID #6–7, 10). Before Patient A began receiving treatment, the patient signed an “Assignment of Benefits” form that assigned their rights under the insurance plan to DaVita. Id. ¶ 31 (Page ID #10). Between April 15, 2017, and August 31, 2018, the costs of Patient A’s dialysis sessions were reimbursed by their health benefit plan, Defendant Marietta Memorial Hospital Employee Health Benefit Plan (the “Plan”), a self-funded plan governed by the Employee Retirement Income Security Act of 1974 (“ERISA”). Id. ¶¶ 13, 29 (Page ID #5, 10). The Plan is funded and administered by Defendant Marietta Memorial Hospital, and its benefit manager is Defendant Medical Benefits Mutual Life Insurance Co. (“MedBen”). Id. ¶¶ 14–15 (Page ID #5).

The Plan provides three tiers of reimbursement benefits, and during the period that Patient A was a member of the Plan, the Plan reimbursed DaVita for the patient’s dialysis costs at the bottom tier, Tier 3. Id. ¶ 24 (Page ID #8). This bottom tier applied to providers, like DaVita, who are “out-of-network.” Id. DaVita was not alone as a bottom-tier dialysis provider—under the Plan’s terms, all dialysis providers are considered out-of-network and are thus subject to lower reimbursement amounts than providers in Tier 1 and Tier 2 are. Id. ¶ 25 (Page ID #8). In addition to this categorically lower reimbursement level, dialysis providers like DaVita are subject to a further, unique limitation. Whereas most out-of-network providers are reimbursed in the bottom tier based on a “reasonable and customary” fee as the term is understood in the healthcare industry, dialysis providers are subject to an “alternative basis for payment.” Id. ¶ 27 (Page ID #9). Specifically, reimbursement for dialysis providers “will not exceed the maximum payable amount applicable . . . which is typically one hundred twenty-five percent (125%) of the current Medicare allowable fee.” Id. (quoting Compl. Ex. A at 17). Finally, for the dialysis service itself, the Plan reimburses at a rate of 70% of the 125% of the Medicare allowable fee—in other words, 87.5% of the Medicare rate, see Appellant Br. at 20—a fee which is already lower than the industry-wide definition of a “reasonable and customary” fee. R. 1 (Compl. ¶ 28) (Page ID #9–10). For these reasons, DaVita was reimbursed at a relatively lower rate both compared to in-network providers and to other out-of-network providers. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 4

DaVita was not the only entity that allegedly suffered due to the Plan’s differential treatment of dialysis reimbursement. During the time that Patient A was covered by the Plan, the patient had no in-network options for dialysis services, exposing them to higher copayments, coinsurance amounts, and deductibles. Id. ¶ 48 (Page ID #16–17). Patient A was allegedly at risk of DaVita billing them for the balance of what the Plan had not reimbursed DaVita. Id. ¶ 35 (Page ID #11–12). The Plan also identified dialysis as subject to heightened scrutiny, such as “cost containment review” and “claim audit and/or review,” which allegedly incentivizes dialysis patients to abandon the Plan and switch to Medicare. Id. ¶ 51 (Page ID #18). On August 31, 2018, Patient A dropped the Plan as a primary insurance provider and switched to Medicare, to which they were entitled by virtue of having ESRD. Id. ¶ 29 (Page ID #10); see 42 U.S.C. § 426–1.

On December 19, 2018, DaVita filed a complaint against the defendants, alleging that the Plan treats dialysis providers differently from other medical providers in violation of the Medicare Secondary Payer Act (“MSPA”) and ERISA. R. 1 (Compl. at 1) (Page ID #1). The complaint is brought on DaVita’s behalf and on behalf of Patient A. Id. ¶ 10 (Page ID #4). The crux of DaVita’s complaint is that by offering inferior benefits to individuals with ESRD, the Plan unlawfully incentivized such individuals, like Patient A, to drop the Plan as their health insurer and go on Medicare. Id. ¶ 6, 50 (Page ID #3, 17–18). On February 14, 2019, Marietta Memorial Hospital and the Plan moved to dismiss DaVita’s complaint, R. 17 (Marietta Mot. to Dismiss at 1) (Page ID #182), and on the next day, MedBen did the same, R. 18 (MedBen Mot. to Dismiss at 1) (Page ID #192).

On September 20, 2019, the district court granted the defendants’ motions to dismiss in a written opinion and order, dismissing all of DaVita’s counts with prejudice. See DaVita, Inc. v. Marietta Mem’l Hosp. Emp. Health Benefit Plan, No. 2:18-CV-1739, 2019 WL 4574500, at *5, 7 (S.D. Ohio Sept. 20, 2019). As to DaVita’s MSPA claim, the district court explained that the MSPA private cause of action was available only to sue for recovery of payments that Medicare had made to a provider when a plan failed to make these payments and that, in this case, Medicare had made no such payments. Id. at *2–3. Alternatively, the district court held that even if this private cause of action were available, the Plan had not discriminated unlawfully No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 5

against individuals with ESRD through its reimbursement system. Id. at *3–5. As to DaVita’s ERISA claims that were premised on MSPA violations, the district court held that its analysis regarding the lack of unlawful discrimination foreclosed these claims. Id. at *5. Lastly, as to DaVita’s ERISA claims that were premised on a breach-of-fiduciary-duty theory, the district court held that the Assignment of Benefits signed by Patient A was insufficient to confer standing on DaVita to bring these claims. Id. at *5–7.2 DaVita timely appealed. R. 48 (Notice of Appeal at 1) (Page ID #494).

II. STANDARD OF REVIEW

We review de novo the district court’s grant of a motion to dismiss. Milligan v. United States, 670 F.3d 686, 692 (6th Cir. 2012). A complaint must include “a short and plain statement of the claim showing that the pleader is entitled to relief,” Fed. R. Civ. P. 8(a)(2), and is properly dismissed if it “fail[s] to state a claim upon which relief can be granted,” Fed. R. Civ. P. 12(b)(6). “[A] federal court of appeals is not restricted to ruling on the district court’s reasoning, and may affirm a district court’s grant of a motion to dismiss on a basis not mentioned in the district court’s opinion.” In re Comshare, Inc. Sec. Litig., 183 F.3d 542, 547–48 (6th Cir. 1999). “This Court reviews de novo the district court’s decision to dismiss a claim for lack of standing.” Coyne v. Am. Tobacco Co., 183 F.3d 488, 492 (6th Cir. 1999).

III. DISCUSSION

A. The Private Cause of Action

DaVita’s complaint first alleges a violation of the MSPA through the Act’s private cause of action. The defendants argue that a Medicare conditional payment is required before a party can sue under this cause of action and that, in this case, Medicare was not allegedly forced to make a conditional payment to cover the Plan’s obligations. We agree with the first proposition

2The district court’s opinion states conflicting bases for dismissing Count VII. Compare DaVita, 2019 WL 4574500, at *5 (appearing to dismiss Count VII on its merits, i.e., failure to allege a violation of the MSPA), with id. at *7 (“Counts Three through Seven are DISMISSED with prejudice, for lack of standing.”) (emphasis added). This conflict is irrelevant, however, given that we review the dismissal de novo whether it was based on the merits or on standing. See supra Part II. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 6

and disagree with the second. Although the plain text of the private-cause-of-action provision requires Medicare to make a conditional payment before the cause of action is available, Medicare allegedly made such a payment in this case. In order to explain why, a brief background on the statutory framework is necessary.3

1. Statutory Background

In 1972, Congress extended Medicare coverage to nearly all individuals with ESRD. See Pub. L. No. 92–603, § 299I, 86 Stat. 1429, 1463–64 (1972). This meant that Medicare served as the primary payer of health costs for such individuals, as well as the many other individuals eligible for Medicare. Amid rising healthcare costs, in 1980, Congress began its effort to counteract escalating Medicare costs by enacting the MSPA, which initially focused on getting private health insurance plans to help cover costs that Medicare had been paying out for automobile accidents.4 See United States v. Baxter Int’l, Inc., 345 F.3d 866, 889 (11th Cir. 2003). In enacting the MSPA, Congress determined that Medicare would have secondary rather than primary liability for accident-injury healthcare costs incurred by a Medicare-eligible individual who had automobile insurance that covered these costs. See H.R. REP. No. 96–1167, at 389 (1980). Thus, if a certain individual was eligible for Medicare but had private insurance, Medicare would serve not as the primary payer but as the secondary payer (i.e., the backup payer) in case the private insurer did not pay. See Stalley v. Methodist Healthcare, 517 F.3d 911, 915 (6th Cir. 2008). If it was reasonably expected that a private health plan would not pay a healthcare provider promptly, Medicare could make a conditional payment to the provider on the plan’s behalf and later seek reimbursement. Id.

In the Omnibus Reconciliation Act of 1981 (“1981 OBRA”), Congress amended the MSPA to cover individuals with ESRD who are on private health insurance. See Omnibus Budget Reconciliation Act of 1981, Pub. L. No. 97-35, § 2146 (1981). With this amendment, an

3An appendix of the various statutes and regulations discussed in this opinion appears at the end of this opinion. 4Prior to 1980, workers’ compensation and other government benefits were also primary payers before Medicare. See Social Security Amendments of 1965, Pub. L. No. 89–97, § 1862(a). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 7

individual’s private insurer—such as an employer-funded health plan, like the Plan in this case— was required to pay for a patient’s dialysis treatment for a period of twelve months5, after which Medicare would take over and become the primary payer for these treatments. The Senate Report for the 1981 OBRA explained that before the amendment, “most health plans . . . contain[ed] provisions that [we]re intended to prevent payment of benefits where the insured [wa]s also entitled to benefits as a result of coverage under a program such as [M]edicare.” S. REP. 97-139, at 735 (1981). The amended Act prohibited this. It denied a health plan’s tax deduction “if the plan differentiates in the benefits it provides between individuals having [ESRD] and other individuals covered by such plan on the basis of the existence of [ESRD], the need for renal dialysis, or in any other manner.” Pub. L. No. 97-35, § 2146(b). As we have observed, “[i]t would appear that the precise problem that Congress sought to ameliorate was that private plans would provide inferior benefits or coverage for medical treatment that also was covered by Medicare.” Bio-Med. Applications of Tenn., Inc. v. Cent. States Se. & Sw. Areas Health & Welfare Fund, 656 F.3d 277, 281 (6th Cir. 2011).

The resulting statute, after several decades of amendments, thus prevents a “group health plan” from taking two actions with respect to an individual diagnosed with ESRD:

A group health plan . . . — (i) may not take into account that an individual is entitled to or eligible for [Medicare benefits due to ESRD] during the [30]-month period which begins with the first month in which the individual becomes entitled to benefits . . . ; and (ii) may not differentiate in the benefits it provides between individuals having end stage renal disease and other individuals covered by such plan on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner[.]

42 U.S.C. § 1395y(b)(1)(C). If a private plan violates these antidiscrimination provisions and fails to pay for a plan member’s dialysis treatment because that individual has become eligible for Medicare benefits by virtue of having ESRD, Medicare still stands in as a backup payer. See Bio-Med., 656 F.3d at 283–84. In practice, this secondary payment system allows individuals

5This period is now thirty months. 42 U.S.C. § 1395y(b)(1)(C) (2016). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 8

with ESRD to continue receiving treatment from a dialysis provider when their primary plan has not paid or is not expected to pay the provider. Specifically, the statute states that when a primary plan “cannot reasonably be expected” to pay “promptly,” Medicare may make “conditional payment[s]” to the provider. 42 U.S.C. § 1395y(b)(2)(B)(i). The statute requires the primary plan to reimburse Medicare for such conditional payments. Id. § 1395y(b)(2)(B)(ii). By contrast, if the primary plan has paid or is reasonably expected to pay the provider, Medicare may not make payments to the provider. See id. § 1395y(b)(2)(A)(i) (prohibiting Medicare from paying as a secondary payer “except as provided in subparagraph (B),” the conditional payment provision).

The MSPA has two internal enforcement mechanisms. The first is the government- enforcement provision, which reads:

In order to recover payment made under this subchapter for an item or service, the United States may bring an action against any or all entities that are or were required or responsible (directly, as an insurer or self-insurer, as a third-party administrator, as an employer that sponsors or contributes to a group health plan, or large group health plan, or otherwise) to make payment with respect to the same item or service (or any portion thereof) under a primary plan.

42 U.S.C. § 1395y(b)(2)(B)(iii). The second is the private-cause-of-action provision, which reads:

There is established a private cause of action for damages (which shall be in an amount double the amount otherwise provided) in the case of a primary plan which fails to provide for primary payment (or appropriate reimbursement) in accordance with paragraphs (1) and (2)(A).

42 U.S.C. § 1395y(b)(3)(A). In the private-cause-of-action provision, paragraph (1) refers to the MSPA’s antidiscrimination provision, and paragraph (2)(A) refers to the MSPA’s Medicare- secondary-payment provision. As we recognized in Bio-Medical, the private-cause-of-action provision is worded somewhat oddly. See 656 F.3d at 279 (describing the “tortuous text” of the MSPA). In particular, the provision suggests that a primary plan could fail to comply with a rule—set forth in subparagraph (2)(A)—that applies to Medicare, not to the primary plan. See id. at 286 (“How can a primary plan fail to make a payment in accordance with subparagraph (2)(A), if that subparagraph only instructs when Medicare, and not primary plans, may or may No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 9

not make payments? The answer, of course, is that it cannot: it is impossible for one to violate an order addressed only to someone else.”). Our solution in Bio-Medical for rendering the private-cause-of-action provision functional was to “consider paragraphs (1) and (2)(A) collectively, rather than individually.” Id. Namely, we explained that “a primary plan fails to pay ‘in accordance with paragraphs (1) and (2)(A)’ when it terminates a planholder’s coverage and thereby induces Medicare to make a conditional payment on its behalf—that is, when the primary plan violates the statutory system that these two paragraphs set into motion.” Id. “Put differently, a primary plan is liable under the private cause of action when it discriminates against planholders on the basis of their Medicare eligibility and therefore causes Medicare to step in and (temporarily) foot the bill.” Id.

2. Availability of the Private Cause of Action When Medicare Has Not Made a Conditional Payment

The defendants argue that, based in part on the above language from Bio-Medical, DaVita is prohibited from suing under the private cause of action because Medicare did not make a conditional payment in this case. DaVita essentially has three responses. First, Bio-Medical does not hold that a payment by Medicare is a precondition to suing under the private cause of action. Second, the text and purpose of the MSPA counsel against requiring this as a precondition. Third, even if payment by Medicare were required, DaVita’s complaint satisfies this requirement because the Plan induced Patient A to enroll in Medicare, which led to Medicare making payments for Patient A’s dialysis before the relevant thirty-month period was over. We hold as follows: (1) Bio-Medical did not hold that a precondition was required, (2) but our independent review of the statutes confirms that it is; (3) however, Medicare’s payment in this case was conditional, and thus the private cause of action is available to DaVita.

a. Bio-Medical does not control this case

Bio-Medical’s thoughtful discussion on the availability of the private cause of action was not necessary to its decision. In Bio-Medical, the panel was faced with a straightforward case of a primary plan’s terminating coverage and Medicare’s stepping in to cover the primary plan’s obligations. The primary plan had completely ceased payments to the dialysis provider upon No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 10

learning that the relevant patient was eligible for Medicare, even though the patient was still a member of the plan. Id. at 280. Given that the primary plan was failing to meet its payment obligations under the MSPA, the dialysis provider turned to Medicare, which began making payments to enable the patient to continue to receive treatment. Id. Thus, the question before the panel was not whether Medicare payments were a precondition for suit under the private cause of action, but whether this cause of action could be used if the primary plan’s payment had not been previously “demonstrated,” such as by a judgment or settlement. Id. at 279.

Despite the clarity of Bio-Medical’s language regarding Medicare “step[ping] in and (temporarily) foot[ing] the bill,” this statement was not necessary to its decision. Instead, this statement is a clear example of dicta. See Black’s Law Dictionary (11th ed. 2019) (defining obiter dictum as “[a] judicial comment made while delivering a judicial opinion, but one that is unnecessary to the decision in the case and therefore not precedential”). This is not a case in which one panel has issued a clear holding on an issue and a litigant in a subsequent case is bound by that holding despite presenting a novel argument that the prior panel did not actively consider in issuing its holding. See, e.g., Grundy Mining Co. v. Flynn, 353 F.3d 467, 479 (6th Cir. 2003); see also United States v. Bawgus, 782 F. App’x 408, 412 n.3 (6th Cir. 2019) (Moore, J., concurring in the judgment). Nor can the relevant statements in Bio-Medical be considered one of several alternative holdings, which standing alone could be viewed as technically unnecessary given the existence of the others. Cf. Woods v. Interstate Realty Co., 337 U.S. 535, 537 (1949) (“[W]here a decision rests on two or more grounds, none can be relegated to the category of obiter dictum.”). The statement that the private cause of action is permitted when Medicare has made a payment was not necessary to the decision, and is thus dicta. 6 Indeed, the

6The Bio-Medical panel’s identification of the “counterargument” to its interpretation of the statute supports the conclusion that the “step[ping] in” language in Bio-Medical was dicta. 656 F.3d at 287. Had the panel held that the text required Medicare payments as a prerequisite for suing under the private cause of action, one might expect the “counterargument” to be that such payments are not, in fact, required. Instead, the panel explained, the “counterargument” to its holding was that “in order to be liable under the Act’s private cause of action, a primary plan’s responsibility to pay must have already been ‘demonstrated’ prior to the lawsuit.” Id. Thus, the discussion regarding Medicare’s temporarily footing the bill was relevant only insofar as it rejected the proposition that a “demonstrated responsibility” was a prerequisite to suit under the private cause of action. Medicare having stepped in and made payments—as opposed to the existence of a “demonstrated responsibility” by a primary plan—was sufficient, but not clearly necessary, for permitting suit under the provision. See Reply Br. at 6 (“To resolve the No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 11

Bio-Medical panel itself did not consider its “step[ping] in” discussion to constitute a holding. Its “Summary of Holdings and Implications for This Case” makes no reference to a requirement that Medicare step in and make a payment to the provider before a private cause of action will be available. 656 F.3d at 294.

b. The MSPA’s private-cause-of-action provision requires a conditional payment by Medicare

Although Bio-Medical does not dictate the outcome in this case, we conclude that the MSPA does require a conditional payment by Medicare before a planholder (or a planholder’s assignee) may sue under the private cause of action. Beginning with the text of the provision, the question is what it means for a primary plan to “fail to provide for primary payment (or appropriate reimbursement) in accordance with paragraphs (1) and (2)(A),” 42 U.S.C. § 1395y(b)(3)(A), or more specifically, whether “in accordance with . . . (2)(A)” constitutes a conditional-payment requirement. As discussed above, paragraph (1) refers to the antidiscrimination provisions of the MSPA, and paragraph (2)(A) refers to the “Medicare secondary payer” provision. The private-cause-of-action provision is explicitly limited to situations in which a primary plan has failed to act in accordance with both paragraphs (1) and (2)(A). See Bio-Med., 656 F.3d at 285 (emphasizing the conjunctive nature of the private-cause- of-action provision). As we said in Bio-Medical, incorporating paragraph (2)(B) into paragraph (2)(A) results in the following formulation of paragraph (2)(A): “Medicare may not pay for any item or service to the extent that the Act requires a primary plan to pay, except that Medicare may conditionally pay for the item or service if the primary plan cannot reasonably be expected to pay promptly.” Id. at 286. The question is how a primary plan might fail to act in accordance with this provision.

In our view, the only way that a primary plan can fail to act in accordance with this provision is by failing to make payments or appropriate reimbursements to a provider and thus triggering the remission of a conditional payment by Medicare. We thus make a holding of our observation in Bio-Medical: “[A] primary plan is liable under the private cause of action when it

liability question, this Court decided only that the payment by Medicare in that case was sufficient to allow the private suit, not that a payment by Medicare was necessary in all circumstances.”). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 12

discriminates against planholders on the basis of their Medicare eligibility and therefore causes Medicare to step in and (temporarily) foot the bill.” 656 F.3d at 286. When we “consider paragraphs (1) and (2)(A) collectively, rather than individually,” discriminating against a planholder in violation of paragraph (1) is what sets in motion the conditional-payment mechanism set forth in paragraph (2). Id. And conditional payment is the only type of payment permitted under paragraph (2)(A). Thus, the only way that a primary plan fails to act in accordance with paragraph (2)(A)—or, in other words, the only way it sets paragraph (2)(A)’s payment mechanism in motion—is by triggering Medicare to make a conditional payment. It is true that, unlike the government-enforcement provision, which explicitly states that enforcement actions may be taken “to recover payment made,” 42 U.S.C. § 1395y(b)(2)(B)(iii), the private- cause-of-action provision does not explicitly identify payment recovery as the purpose of the provision. Appellant Br. at 29–30; Reply Br. at 4–5. But the private-cause-of-action provision’s reference to paragraph (2)(A) unmistakably incorporates a prerequisite that Medicare has paid the provider. And as for DaVita’s purpose-based argument about the private-cause-of-action provision, “[c]ourts considering the provision have generally agreed that the apparent purpose of the statute is to help the government recover conditional payments from insurers or other primary payers.” Stalley v. Cath. Health Initiatives, 509 F.3d 517, 524 (8th Cir. 2007).

An alternative reading of the private-cause-of-action provision—one that, we note, no party has raised—is that the provision requires the showing of a Medicare payment, but not a conditional Medicare payment, as a precondition to suit. Under this theory, a plan would fail to pay (or appropriately reimburse) in accordance with paragraph (2)(A) when it forces Medicare to break with the paragraph’s general prohibition on payments other than conditional ones. That is, because paragraph (2)(A) contemplates only conditional payments, a plan that induced Medicare to make a payment outside what the statute has contemplated would contravene paragraph (2)(A). The problem with this interpretation, however, is that, per the plain text of paragraph (2), the only type of payment that Medicare is capable of making as a secondary payer is a conditional one. Conditional payments are not simply the kinds of payments Medicare may make under certain circumstances; they are the only kind of payments that Medicare can make as a secondary payer. Put another way, Medicare stepping in to make a payment when a plan has No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 13

failed to do so is, by definition, conditional on the plan ultimately repaying it. The premise of the alternative reading—that Medicare can be forced into making payments that exceed the scope of its statutory authority—finds no support in the text. Thus, to act in contravention of paragraphs (1) and (2)(A) is to violate the first and set in motion the conditional-payment mechanism of the second.

Resisting this conclusion, Amicus Dialysis Patient Citizens (“DPC”) maintains that we should focus on the disjunctive nature of “primary payment (or appropriate reimbursement)” in the private-cause-of-action provision. 42 U.S.C. § 1395y(b)(3)(A) (emphasis added). In DPC’s view, “[t]he statute envisions two alternate scenarios; one where Medicare has made a conditional payment and the provider sues for ‘appropriate reimbursement’ to Medicare and another where the private insurer simply ‘fails to provide primary payment’ to the provider.” DPC Amicus Br. at 25. But there is an equally compelling interpretation of this disjunctive language that incorporates the text’s requirement of a conditional payment: Medicare conditional payments are anticipated in both scenarios, and the provision simply allows for private suits at different stages of the payment/reimbursement process. The “fail[ure] to provide for primary payment” scenario would involve a primary plan failing to make payments, Medicare stepping in to make these payments conditionally, and a private party suing for the amount that Medicare has paid. The “fail[ure] to provide . . . appropriate reimbursement” scenario, by contrast, would involve a primary plan failing to make payments, Medicare stepping in to make these payments conditionally, the primary plan remitting insufficient reimbursement to Medicare, and the private party suing for the additional amount that the primary plan should have remitted to Medicare. This interpretation would give independent meaning to both parts of the statutory phrase, and would comport with the text’s requirement of a conditional payment by Medicare.

c. DaVita has plausibly alleged that Medicare made a conditional payment

DaVita suggests that the conclusion we have laid out above will make it impossible for private actors to sue for discrimination. According to DaVita, the MSPA’s extremely high threshold for allowing Medicare to make a conditional payment—barring payment unless the No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 14

primary plan “has not made or cannot reasonably be expected to make” a payment—makes conditional payments exceedingly rare. If we were to conceive of conditional payments as narrowly as DaVita suggests, DaVita would undoubtedly be correct that a “perverse result” would ensue. Reply Br. at 5. Indeed, a narrow understanding of conditional payments would permit precisely the type of discrimination that the MSPA prohibits. This is a result that neither the text nor purpose of the MSPA can bear. Taking the allegations in DaVita’s complaint to be true, Medicare has allegedly made a conditional payment in this case.7

Our task is to determine when Medicare can make conditional payments. The “[a]uthority to make conditional payment” section of the MSPA provides that “[t]he Secretary may make payment under this subchapter with respect to an item or service if a primary plan . . . has not made or cannot reasonably be expected to make payment with respect to such item or service promptly.” 42 U.S.C. § 1395y(b)(2)(B)(i). One option is to construe the phrase “has not made or cannot reasonably be expected to make payment” to mean “has not made or cannot reasonably be expected to make any payment at all.” Under this approach, the only scenario in which a planholder (or its assignee) could sue for unlawful discrimination under the MSPA is when (1) a primary plan terminated the planholder’s coverage completely or flatly refused to make any further payments to the provider, and (2) Medicare stepped in to make these payments. See, e.g., Bio-Med., 656 F.3d at 280 (Medicare paid provider after primary plan terminated a patient’s coverage upon learning that the patient was eligible for Medicare benefits). In other words, so long as the plan has paid the provider something for a given service, any gap-filling payment by Medicare would not be considered conditional. But this would “thereby insulate [the plan’s] actions from any judicial review” under the MSPA itself. Appellant Br. at 31. Construing the statute in this way would run directly contrary to the text of the MSPA—which prohibits discrimination against individuals with ESRD, as discussed below—and its purposes of protecting both ESRD patients and Medicare’s fiscal integrity.

Instead, we construe the conditional-payment requirement in light of its neighboring provisions, which prohibit a plan from “tak[ing] into account” an individual’s Medicare

7DaVitaraises this argument in the alternative to its argument that the MSPA does not require a conditional payment. See Appellant Br. at 36–40. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 15

entitlement or discriminating against an individual for having ESRD. 42 U.S.C. § 1395y(b)(1)(C). A plan may violate these proscriptions not only by terminating a patient’s coverage completely, but also by providing differential benefits; either form of discrimination would result in insufficient payment to a provider, and in turn, the possibility of a conditional payment by Medicare to ensure that the provider continues providing treatment to the patient. It is true that DaVita did not label the alleged payments by Medicare as “conditional” in its complaint. R. 1 (Compl. ¶¶ 1–94) (Page ID #1–29). But if DaVita’s allegations prove true, then the only difference between this case and a termination-of-coverage case like Bio-Medical, for example, is the level of sophistication used by the primary plan in kicking a patient off coverage and forcing the patient’s costs onto Medicare. If the allegations are true, then, in essence, Medicare’s payments were made as a secondary payer, conditioned on the Plan’s repayment after initially failing to abide by its primary-payer duties under the MSPA. Thus, where MedBen asserts, “Once Patient A left the Plan, Medicare was the primary payer,” MedBen Br. at 3, we respond: Not if the Plan’s unlawful actions forced Medicare into the role of primary payer in the first place.

This is not to say that anytime Medicare remits payment to a provider, the provider or planholder will have a private cause of action under the MSPA. Medicare makes payments to providers all the time that are plainly not “conditional payments.” For example, Medicare serves as the primary payer for individuals who do not have private health insurance at all, and thus makes regular payments on their behalf to providers. See 42 U.S.C. § 426-1(a)(2) (setting forth entitlement to Medicare benefits for those who are “medically determined to have end stage renal disease”). Furthermore, if an individual who makes a truly voluntary decision to drop their private health insurance during the thirty-month coordination period and join Medicare, subsequent payments by Medicare to this individual’s providers are not conditional in any sense. Instead, by virtue of the individual’s choice, Medicare would be the primary payer and pay accordingly. But that is not the situation we have here. In contrast to these hypotheticals, DaVita alleges that Patient A’s decision to leave the Plan and go on Medicare was not truly voluntary; it was instead a direct result of the Plan’s discriminatory underpayment. Conditional payments do arise in such situations. If DaVita prevails in this lawsuit, demonstrating that No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 16

Medicare made payments that it never should have made, the Plan should be obligated to repay Medicare per 42 U.S.C. § 1395y(b)(2)(B)(i).

We acknowledge that one implementing regulation appears to prevent conditional payment by Medicare when a plan is engaged in discriminatory underpayment. See Appellant Br. at 30, Reply Br. at 5. Section 411.165 of the Code of Federal Regulations provides that Medicare “does not make conditional primary payments” when a “[group health] plan denies a claim in whole or in part” because the “plan limits its payments when the individual is entitled to Medicare.” 42 U.S.C. § 411.165. Consider a scenario where a plan’s benefits structure partially denies all dialysis claims from ESRD patients by reimbursing providers at 0.01% of the billed rate due to these patients’ entitlement to Medicare. Applied literally, the regulation would prohibit Medicare from filling in the enormous gap in payment while the patient is still enrolled in the plan and bar Medicare from footing the bill if the patient involuntarily leaves the plan. In effect, the regulation extinguishes the non-differentiation and the take-into-account clauses’ protections throughout the entirety of the thirty-month coordination period.

In our view, this regulation’s interpretation of conditional payment and the surrounding statutory framework conflicts both with the text of the MSPA and other implementing regulations. See 42 U.S.C. § 1395y(b)(1)(C)(i) (“A group health plan . . . may not take into account that an individual is entitled to or eligible for [Medicare] benefits.”); 42 C.F.R. § 411.108(a)(5) (“Actions by [group health plans] or [large group health plans] that constitute taking into account that an individual is entitled to Medicare on the basis of ESRD . . . include, but are not limited to, the following: . . . Imposing limitations on benefits for a Medicare entitled individual that do not apply to others enrolled in the plan, such as providing less comprehensive health care coverage, excluding benefits, reducing benefits, charging higher deductibles or coinsurance, providing for lower annual or lifetime benefit limits, or more restrictive pre-existing illness limitations.”). Deference to agency interpretation of a statute is in order only when such interpretation is “not in conflict with the plain language of the statute.” Kmart Corp. v. Cartier, Inc., 486 U.S. 281, 292 (1988); Garcia v. Sec’y of Health & Hum. Servs., 46 F.3d 552, 557 (6th Cir. 1995) (“[W]here the Secretary’s interpretation conflicts with the plain language of the governing statute, we will not hesitate to overturn that interpretation.”). By way of narrowing No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 17

the availability of conditional payments, the regulation condones plan differentiation in benefits based on an individual’s entitlement to Medicare—an interpretation that is directly at odds with the MSPA’s text. Accordingly, we do not defer to such an interpretation.

Permitting suit under the private cause of action in this situation, when Medicare allegedly began paying before it should have, also comports with the double-damages nature of the private cause of action. See 42 U.S.C. § 1395y(b)(3)(A) (“[D]amages . . . shall be in an amount double the amount otherwise provided.”). We have explained that apart from promoting a general interest in punishment and deterrence for misconduct by primary plans, double damages serve the purposes of incentivizing private parties to bring lawsuits while also recouping Medicare’s financial losses. See Bio-Med., 656 F.3d at 279; Star Dialysis, LLC v. WinCo Foods Emp. Benefit Plan, 401 F. Supp. 3d 1113, 1128 (D. Idaho 2019) (“The statute provides for double damages to allow Medicare to recoup any conditional payments, and to offer a reward to the private litigant bringing the action.”). If DaVita is able to demonstrate that, as a result of the Plan’s insufficient payment or reimbursement, Patient A was forced off of the Plan on August 31, 2018, leading to sixteen months of conditional payments by Medicare, then the private cause of action would serve to recoup these payments while also rewarding DaVita for bringing the litigation.

For the foregoing reasons, we conclude that the district court erred in dismissing Count I of DaVita’s complaint. We hold that a conditional payment by Medicare is required as a precondition to suing under the MSPA’s private cause of action and that the complaint contains sufficient allegations of such a payment for DaVita to proceed further with Count I.

B. The Assignment of Benefits

Aside from Count I, the remainder of DaVita’s complaint alleges violations of ERISA. Before assessing the merits of these claims, we must answer a threshold question: May DaVita raise these claims through its status as an assignee of Patient A?8 The district court answered

8This is distinct from the question of standing, see Cranpark, Inc. v. Rogers Grp., Inc., 821 F.3d 723, 730 (6th Cir. 2016), which merits only brief discussion in this case. Article III standing requires a plaintiff to have “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 18

“no” as to Counts III through VI and did not address the issue as to Counts II and VII. See DaVita, 2019 WL 4574500, at *5 & n.2.9 MedBen, named as a defendant in Counts II and IV–VI, argues that DaVita’s “equitable ERISA claims”—which correspond to Counts IV–VI, but not II—should be dismissed for lack of standing. MedBen Br. at 48. Marietta and the Plan, named as defendants in Counts II–III and VII,10 argue only that Count III should be dismissed for lack of standing. Marietta Br. at 10.

The putative basis for DaVita’s right to bring the six ERISA claims is the Assignment of Benefits form that Patient A signed prior to receiving treatment from DaVita. The form reads:

I hereby assign to Facility and DaVita all of my right, title and interest in any cause of action and/or any payment due to me (or my estate) under any employee benefit plan, insurance plan, union trust fund, or similar plan (‘Plan’), under which I am a participant or beneficiary, for services, drugs or supplies provided by Facility to me or my dependents for purposes of creating an assignment of benefits under ERISA or any other applicable law. I also hereby designate DaVita as a beneficiary under any such Plan and instruct that any payment be made solely to and sent directly to DaVita. If I receive any payment directly from any Plan for services, drugs or supplies provided to me by DaVita, including insurance checks, I recognize that such payment sent directly to me was

to be redressed by a favorable judicial decision.” Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547 (2016), as revised (May 24, 2016) (citing Lujan v. Defs. of Wildlife, 504 U.S. 555, 560–61 (1992)). Assuming a valid assignment from Patient A to DaVita, DaVita has standing if Patient A suffered an injury in fact. See Springer v. Cleveland Clinic Emp. Health Plan Total Care, 900 F.3d 284, 287–88 (6th Cir. 2018) (“Non-participant health care providers cannot bring their own ERISA claims—they do so derivatively, relying on the participants’ contractually defined rights and therefore the participants’ standing at the time of the assignment.”). The complaint clearly alleges injuries in fact, describing Patient A’s exposure to “higher copayments, coinsurance amounts, and/or deductibles.” R. 1 (Compl. ¶ 48) (Page ID #16–17). These risks allegedly arose due to the Plan’s denial of benefits, which Patient A was entitled to receive in a nondiscriminatory manner by law. As our decision in Springer makes clear, “injury does not depend on allegation of financial loss.” 900 F.3d at 287. In Springer, as here, the provider—not the participant— had incurred a financial loss due to underpayment by the primary plan. The principal difference between Springer and this case—Springer’s provider was allegedly underpaid under the terms of the health plan, whereas here, the provider was allegedly underpaid under the terms of the MSPA—does not render Springer inapposite. In both cases, the participants were allegedly denied benefits, even if they “were never at imminent risk of out-of-pocket expenses.” Id. (quoting N. Cypress Med. Ctr. Operating Co. v. Cigna Healthcare, 781 F.3d 182, 192 (5th Cir. 2015)). The fact that DaVita did not bill Patient A for the amount that it believes it was underpaid does not alter this conclusion. See Spinedex Physical Therapy USA Inc. v. United Healthcare of Ariz., Inc., 770 F.3d 1282, 1289 (9th Cir. 2014) (rejecting the proposition that “because [the provider] has not sought payment from its assigning patients for any shortfall, those patients do not have the ‘injury in fact’ necessary for Article III standing”). Accordingly, DaVita has Article III standing to bring its ERISA claims. 9But see supra note 2. 10For Count III, only Marietta is named as a defendant. See R. 1 (Compl. ¶¶ 71–73) (Page ID #24–25). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 19

inappropriate and I agree to immediately endorse and forward such payment to DaVita.

R. 1 (Compl. ¶ 31) (Page ID #10–11). This form clearly confers a right on DaVita to bring Count II, and the defendants do not argue otherwise.11 Count II is a claim for unpaid benefits, inter alia, under 29 U.S.C. § 1132(a)(1)(B) of ERISA; it specifically alleges that the defendants “fail[ed] to pay DaVita what they were obligated to pay.” Id. ¶ 69–70 (Page ID #23–24). The form undoubtedly assigns Patient A’s rights as a beneficiary under ERISA to DaVita: “I hereby assign . . . all of my right, title[,] and interest in any cause of action and/or any payment due to me . . . under any [plan], under which I am a participant or beneficiary, for services . . . provided by [DaVita] . . . for purposes of creating an assignment of benefits under ERISA. . . . I also hereby designate DaVita as a beneficiary under any such Plan . . . .” R. 1 (Compl. ¶ 31) (Page ID #10–11) (emphasis added).12

11The only argument that any of the defendants makes with respect to standing and Count II is that DaVita failed to plead an injury by Patient A and therefore lacks standing. See Marietta Br. at 9–10. This is incorrect for reasons discussed above. See supra note 8. 12The concurrence raises a new argument that neither party submitted to the district court, that the district court did not consider in its decision, and that neither party briefed. The concurrence submits that DaVita cannot enforce the MSPA through § 1132(a)(1)(B) because the provision merely “allows plan participants to sue to enforce their rights under a plan’s terms” and does “not allow them to invalidate those terms[.]” Concurring Op. at 52. “A nearby paragraph—§ 1132(a)(3)—. . . [gives] parties a cause of action to challenge a plan’s legality.” Id. at 53. Yes, §§ 1132(a)(1)(B) and 1132(a)(3) are distinct. “[Section 1132(a)(1)(B)] speaks of ‘enforc[ing]’ the ‘terms of the plan,’ not of changing them.” CIGNA Corp. v. Amara, 563 U.S. 421, 436 (2011) (alteration in original) (citing 29 U.S.C. § 1132(a)(1)(B)). A common § 1132(a)(1)(B) challenge involves the following scenario: a plan—in contravention of its own specifications or an employer’s promises—improperly denies benefits to a claimant. In these situations, the proper, and logical, remedy is a court’s invoking § 1132(a)(1)(B) to enforce a plan as written. Section 1132(a)(3), on the other hand, provides that a “participant, beneficiary, or fiduciary” may seek “to enjoin any practice or act which violates [ERISA]” or “to obtain other appropriate equitable relief[,]” 29 U.S.C. § 1132(a)(3), such as plan reformation, monetary compensation, and equitable estoppel, see Amara, 563 U.S. at 441–42. But the two provisions are not oil and water. Courts may first reform a plan’s terms per § 1132(a)(3) before proceeding to enforce the reformed plan per § 1132(a)(1)(B). See Amara, 563 U.S. at 435, 438–40; see also Laurent v. PricewaterhouseCoopers LLP, 945 F.3d 739, 747–49 (2d Cir. 2019) (describing plan reformation under § 1132(a)(3) and subsequent enforcement of the reformed plan under § 1132(a)(1)(B) as a two-step process). In the present case, DaVita alleges that the as-written Plan is illegal. DaVita does not want to—nor should a court—enforce a purportedly illegal plan; instead, DaVita seeks an equitable remedy. In its complaint, under “Count II,” DaVita asks the district court to “sever[]” the “terms [of the Plan] that violate federal law” and “reimburse DaVita pursuant to the terms of the Plan document and other applicable law.” R. 1 (Compl. ¶ 66) (Page ID #22). Accordingly, DaVita seeks benefits that Patient A would have received under a lawful version of the Plan. A court’s awarding Patient A’s unpaid benefits per the as-written Plan would not change the fact that the Plan may No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 20

As to Counts III through VI, by contrast, we conclude that DaVita lacks an assigned right or interest in them. DaVita raises these four breach-of-fiduciary-duty claims under § 1104(a)(1)(B) and quotes the assignment selectively to argue that “Patient A assigned ‘any

illegally discriminate against those with ESRD. See Hill v. Blue Cross & Blue Shield of Mich., 409 F.3d 710, 718 (6th Cir. 2005). Should DaVita succeed upon remand, “[o]nly injunctive relief of the type available under § 1132(a)(3) will provide the complete relief sought by [DaVita] by requiring [the Plan] to alter the manner in which it administers [dialysis claims].” Id; see also Amara, 563 U.S. at 440 (“[A] maxim of equity states that equity suffers not a right to be without a remedy.”) (internal quotation marks and alterations omitted). Perhaps the concurrence is concerned that DaVita explicitly cited only to § 1132(a)(1)(B) in Count II. But we cannot torpedo DaVita’s claim simply because counsel did not write out “§ 1132(a)(3)” under “Count II” in the complaint. Again, DaVita explicitly seeks equitable relief in Count II in line with § 1132(a)(3). If a missing number unsettles the concurrence, perhaps it would be proper for the district court to allow DaVita to amend Count II to add “§ 1132(a)(3)” to the end of ¶ 66 pursuant to Federal Rule of Civil Procedure 15(a)(2). See FED. R. CIV. P. 15(a)(2). The concurrence adds that it “does not see” how DaVita can “fix this problem” by relying on § 1132(a)(3) because the subsection purportedly “permits equitable relief to remedy ‘a practice which violates any provision of this subchapter,’ not a practice that violates a different law.” Concurring Op. at 53 (citing 29 U.S.C. § 1132(a)(3)). We consider the full text of § 1132(a)(3), which provides that a “participant [or] beneficiary” may bring a civil action “(A) to enjoin any act or practice which violates any provision of this subchapter or the terms of the plan, or (B) to obtain other appropriate equitable relief (i) to redress such violations or (ii) to enforce any provisions of this subchapter or the terms of the plan[.]” 29 U.S.C. § 1132(a)(3) (emphasis added). Here, DaVita does not merely allege that it should be reimbursed more for providing dialysis services because the Plan violates the MSPA; DaVita also complains that the Plan’s “singl[ing] out dialysis services for further reimbursement limitations[,]” R. 1 (Compl. ¶ 27) (Page ID #9), violates other terms of the Plan that reimburse other in-network or out-of-network services at higher rates. Cf. Star Dialysis, 401 F. Supp. 3d at 1140 (permitting DaVita to amend its claim to “alleg[e] a violation of the Plan terms” when DaVita’s complaint “pled only that it was not paid ‘the usual and customary rates for DaVita’s services,’ and that DaVita was paid a ‘small fraction of the usual and customary amounts DaVita charges and receives for its services.’”). Specifically, DaVita complains that the Plan reimburses services received from in-network providers but offers “no network of contracted dialysis providers.” R. 1 (Compl. ¶ 25) (Page ID #8). Further, DaVita alleges, “[t]he Plan generally provides for reimbursement based on a ‘reasonable and customary’ fee if a provider is ‘out-of-network.’” Id. ¶ 26. For most out- of-network services, the “reasonable and customary” fee follows the healthcare industry’s understanding of “reasonable and customary”: “a measure of reimbursement based on providers’ billed charges in a particular geographic area.” Id. (Page ID #8–9). However, per DaVita, dialysis treatments are the Plan’s only out-of-network service subject to an “alternative basis for payment” that is based on a percentage of “the current Medicare allowable fee.” Id. ¶ 27 (Page ID #9); see also id. ¶ 49 (Page ID #17) (“In other words, the [] Plan manipulates the definition of ‘reasonable and customary’ to be based on a percentage of Medicare (contrary to the general industry understanding of usual, customary, and reasonable rates), and does so solely for out-of-network dialysis services.”). To the extent that DaVita asks the district court to reform the Plan to extend its in-network reimbursement rates to dialysis providers or to reimburse dialysis services based on the “reasonable and customary” fee that the Plan uses to reimburse every other out-of-network service, id. ¶ 52 (Page ID #18), DaVita seeks appropriate equitable relief to enforce the terms of the Plan per § 1132(a)(3), id. at ¶ 55 (Page ID #20). Of course, these allegations are intertwined with DaVita’s complaints that the Plan violates the MSPA. Id. ¶ 52 (Page ID #18) (“[T]hese provisions removing dialysis patients’ access to in-network options, drastically reducing reimbursement, and singling out dialysis benefits for heightened scrutiny run afoul of the MSPA[] . . . .”). This overlap is expected, as DaVita, in Count I, relies on the MSPA’s private right of action to sue Defendants. But DaVita’s MSPA claims do not undermine DaVita’s concurrently stating sufficient facts that the Plan’s dialysis reimbursement provisions violate the Plan’s terms for reimbursing in-network and other out-of-network services and do not bar DaVita’s seeking equitable relief to “redress such violations.” 29 U.S.C. § 1132(a)(3). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 21

cause of action’ to DaVita, under ‘ERISA or any other applicable law.’” Appellant Br. at 59. It may not be immediately apparent from the Assignment of Benefits form that Patient A successfully assigned to DaVita their right to seek legal and equitable relief as to unpaid benefits per Count II but failed to assign their right to challenge Defendants’ fiduciary duties per Counts III–VI. But a close examination of the form’s textual jungle reveals the difference. As highlighted above, the form stresses Patient A’s transferring their rights as a beneficiary. The form links “any cause of action” and “any payment due” to the patient’s status as a “beneficiary” and to the patient’s “creating an assignment of benefits under ERISA.” Thus, Patient A did not assign “any cause of action under ERISA or any other applicable law” for any purpose whatsoever; rather this patient assigned causes of action brought to recover benefits. See DaVita, Inc. v. Amy’s Kitchen, Inc., 379 F. Supp. 3d 960, 970 (N.D. Cal. 2019) (examining similar language and concluding that it “suggests that, at most, Patient 1 transferred to DaVita the right to bring suit for payment of benefits, rather than for ‘any cause of action’ whatsoever”). Further, the policy stipulates that Patient A “designate[s] DaVita as a beneficiary under [the] Plan”; nowhere does the form explicitly designate DaVita as an inheritor of Patient A’s fiduciary relationship with Defendants.

DaVita protests that “[i]f the parties intended only for Patient A to assign some causes of action, DaVita and Patient A could have said so,” Appellant Br. at 59–60, but the entirety of the assignment, as discussed, shows that they did. Moreover, the title of the assignment is “Assignment of Benefits,” which informs the issue of the assignment’s scope. DaVita’s final argument is that there must be a distinction between “any cause of action” and “any payment due,” and linking the “any cause of action” language to the recoupment of benefits renders it surplusage. See Appellant Br. at 61; Reply Br. at 29. But “any cause of action” could mean something different from “any payment due” without interpreting the former phrase to mean “any cause of action under any existing law.” For instance, the assignment could mean that DaVita is the assignee of both Patient A’s interest in lawsuits related to payments as well as Patient A’s receipt of payments outside the context of litigation.

As to Count VII, Defendants Marietta and the Plan—the only parties named as defendants for this claim, see R. 1 (Compl. ¶¶ 91–94) (Page ID #29)—have forfeited any No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 22

argument that the assignment does not allow DaVita to bring this claim as an assignee. 13 And unlike Article III standing, this issue is waivable. “Article III ‘standing . . . is jurisdictional and not subject to waiver.’” LPP Mortg., Ltd. v. Brinley, 547 F.3d 643, 647 (6th Cir. 2008) (quoting Lewis v. Casey, 518 U.S. 343, 349 n.1 (1996)). The question of whether Patient A transferred their interest to DaVita, by contrast, deals not with Article III standing but with the real-party-in- interest requirement of Federal Rule of Civil Procedure 17, which states that “[a]n action must be prosecuted in the name of the real party in interest.” Fed. R. Civ. P. 17(a)(1); see Fed. R. Civ. P. 17(a)(1)(A)–(G) (“The following may sue in their own names without joining the person for whose benefit the action is brought: (A) an executor; (B) an administrator; (C) a guardian; (D) a bailee; (E) a trustee of an express trust; (F) a party with whom or in whose name a contract has been made for another’s benefit; and (G) a party authorized by statute.”). We have explained that the distinction between this requirement and Article III standing is “critical . . . because the real-party-in-interest requirement is generally viewed as ‘an affirmative defense that can be waived,’ while Article III standing is plaintiff’s burden to prove and can be raised at any point.” Cranpark, Inc. v. Rogers Grp., Inc., 821 F.3d 723, 730 (6th Cir. 2016) (citation omitted). Challenges under this requirement may be “waived or forfeited.” Id. Because DaVita has Article III standing and because Marietta and the Plan make no argument that DaVita cannot bring Count VII as an assignee, we consider this count as well.

We reject the concurrence’s overly formalistic and novel textual challenge that § 1182(a)(1) “applies to a plan’s rules of eligibility, not to its rules concerning covered benefits.” Concurring Op. at 54. Section 1182(a)(1) covers “rules for eligibility (including continued eligibility) of any individual to enroll under the terms of the plan based on . . . health status- related factors[,]” including a person’s “[m]edical condition[.]” 29 U.S.C. § 1182(a)(1). The regulations governing § 1182 are clear—“rules for eligibility” and “rules for continued eligibility” encompass benefits provisions. See 29 C.F.R. § 2590.702(b)(1)(i) (“A group health plan . . . may not establish any rule for eligibility (including continued eligibility) of any individual to enroll for benefits under the terms of the plan . . . that discriminates based on any

13With respect to Count VII, Marietta and the Plan argue only that DaVita has “fail[ed] to state a claim for [a] violation of 29 U.S.C. §1182(a)(1) in Count VII of the Complaint.” Marietta Br. at 11. These defendants present no argument related to standing or the real-party-in-interest requirement. See id. at 11–12. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 23

health factor that relates to that individual . . . .”). The regulations specify that “rules for eligibility” include “rules relating to” “[e]ligibility for benefit packages”; “[b]enefits (including rules relating to covered benefits, benefit restrictions, and cost-sharing mechanisms such as coinsurance, copayments, and deductibles)”; “[c]ontinued eligibility”; and “[t]erminating coverage (including disenrollment) of any individual under the plan.” Id. § 2590.702(b)(1)(ii). Put simply, rules governing a plan’s benefits are “rules for eligibility.” Logic also dictates that a capacious reading of the section—not the concurrence’s narrow exposition—is appropriate. Consider how a policy’s benefits provisions can operate in a manner that impermissibly affects eligibility or enrollment. Here, DaVita does not merely allege that the Plan discriminates against those with ESRD—DaVita also complains that the Plan’s benefits terms force all persons with ESRD to involuntarily leave the Plan and enroll in Medicare.

A superficial glance at § 1182(a)(2)(B) provides tepid support for an interpretation that excludes rules governing benefits from “rules for eligibility.” See 29 U.S.C. § 1182(a)(2)(B) (“[Section 1182(a)(1)] shall not be construed . . . to prevent such a plan or coverage from establishing limitations or restrictions on the amount, level, extent, or nature of the benefits or coverage for similarly situated individuals enrolled in the plan or coverage.”). But, again, the provision’s accompanying regulations support our understanding that § 1182(a)(1) bars plans from limiting benefits based on a health status-related factor: “a plan may limit or exclude benefits in relation to a specific disease or condition, limit or exclude benefits for certain types of treatments or drugs, or limit or exclude benefits based on a determination of whether the benefits are experimental or not medically necessary, but only if the benefit limitation or exclusion applies uniformly to all similarly situated individuals and is not directed at individual participants or beneficiaries based on any health factor of the participants or beneficiaries.” Id. § 2590.702(b)(2)(i)(B) (emphasis added). The regulations contain countless illustrative examples of how rules that govern a plan’s benefits may or may not violate § 1182, which further No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 24

buttresses our understanding of this provision. See 29 C.F.R. § 2590.702(b)(2)(i)(D) (Examples 1–7).14

C. MSPA Discrimination

Arriving to the merits of Counts II and VII, the basic question is whether the MSPA prohibits primary plans from discriminating against individuals with ESRD without expressly stating that these individuals will be treated differently. If the answer is yes, then DaVita has plausibly alleged two ERISA violations. First, if the defendants discriminated against Patient A in violation of federal law (i.e., the MSPA), then DaVita, as Patient A’s assignee, is entitled to unpaid benefits under ERISA flowing from this violation. See 29 U.S.C. § 1132(a)(1)(B); R. 1 (Compl. ¶¶ 66–67, 70) (Page ID #22–24). Second, if the defendants “establish[ed] rules for eligibility . . . of any individual to enroll under the terms of the [group health] plan” based on an

14The Plan’s allegedly singling out dialysis treatments corresponds with two of the regulations’ examples. Example 2 contemplates the following scenario: a “group health plan has a $500 deductible on all benefits for participants covered under the plan” and a participant files “a claim for the treatment of AIDS.” 29 C.F.R. § 2590.702(b)(2)(i)(D) (Example 2). The plan’s corporate board “discuss[es]” the claim; “[s]hortly thereafter, the plan is modified to impose a $2,000 deductible on benefits for the treatment of AIDS, effective before the beginning of the next plan year.” Id. The regulations “[c]onclu[de]” that the facts of Example 2 “strongly suggest that the plan modification is directed at [the participant] based on [the participant’s AIDS] claim.” Id. Here, DaVita contends that Defendants’ limiting dialysis reimbursements across the board may be driven by the Plan’s “incentive to unload ESRD patients whose chronic illness costs the plan more than their other enrollees.” R. 1 (Compl. ¶ 40) (Page ID #14). Discovery may yield evidence of Defendants’ motive for instituting unique reimbursement terms for dialysis services. DaVita’s allegations also align with the facts of Example 5. This example presents a scenario where a group health plan that “applies a $2 million lifetime limit on all benefits” reduces the limit “to $10,000 for any participant or beneficiary covered under the plan who has a congenital heart defect.” Id. (Example 5). Again, the regulations “conclud[e]” that this would violate § 1182 because “[the] benefits under the plan are not uniformly available to all similarly situated individuals and the plan’s lifetime limit on benefits does not apply uniformly to all similarly situated individuals.” Id. Here, the Plan does not explicitly single out persons with ESRD. But, as we explain in greater detail in Section C., infra, DaVita plausibly alleges that “all, or virtually all, of the enrollees who are affected by this discriminatory provision are Plan members suffering from ESRD.” R. 1 (Compl. ¶ 28) (Page ID #9–10). Essentially, DaVita complains that the Plan operates like the problematic plan described in Example 5. The concurrence skips over Examples 2 and 5 and turns to Example 4 for the proposition that “[t]he regulation addressing this section . . . expressly permits uniform limitations on benefits for certain diseases.” Concurring Op. at 54. Example 4 provides that a group health plan’s placing a “$2,000 lifetime limit for treatment of temporomandibular joint syndrome (TMJ)” that is “applied uniformly to all similarly situated individuals and is not directed at individual participants or beneficiaries” does “not violate paragraph (b)(2)(i).” 29 C.F.R. § 2590.702(b)(2)(i)(D) (Example 4). The present case is distinct because DaVita claims that the dialysis reimbursement rates are directed at individual beneficiaries with ESRD and that the Plan’s benefits do not uniformly apply to similarly situated individuals because the dialysis services reimbursement rates affect only those with ESRD. See R. 1 (Compl. ¶ 52) (Page ID #18). To the extent that Example 4 conflicts with Examples 2 and 5, Example 4 should not be followed. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 25

array of “health status-related factors,” then they violated § 1182(a)(1) of ERISA, and § 1132(a)(3) provides a private right of action to “enjoin any act or practice which violates any provision of this subchapter,” 29 U.S.C. § 1132(a)(3).

We hold that the MSPA’s antidiscrimination provisions prohibit conduct beyond the express differential treatment of individuals with ESRD. First, as to the non-differentiation provision of the MSPA, its plain text prohibits both express anti-ESRD discrimination based on an individual’s ESRD status and indirect anti-ESRD discrimination based on an individual’s ESRD-specific need for renal dialysis or based on any other factor. Second, as to the take-into- account provision, the meaning of “take into account” is ambiguous, but the relevant regulations support DaVita’s theory of discrimination for the same reasons that it states a violation of the non-differentiation provision. In short, a plan may be engaging in unlawful discrimination against individuals with ESRD even if it does not explicitly single these individuals out for differential treatment.

1. The Non-Differentiation Provision

Paragraph (1)(C)(ii) of the MSPA states that a group health plan “may not differentiate in the benefits it provides between individuals having end stage renal disease and other individuals covered by such plan on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner.” 42 U.S.C. § 1395y(b)(1)(C)(ii). The provision thus specifies three different ways in which a plan may unlawfully discriminate against individuals with ESRD. The first is by explicitly fashioning differential benefits for ESRD patients by virtue of them having ESRD. If, for example, the Plan included a provision that all treatments required by Plan participants who had ESRD were considered out-of-network, this would single ESRD patients out based on the “existence” of their ESRD diagnosis. Id.

The second and third types of anti-ESRD discrimination are at issue here. The second is the differential treatment of ESRD patients based on their “need for renal dialysis.” As DaVita explains, “need” includes these patients’ need for the treatment compared to those who do not need it at all, and their need compared to those who do need it, just to a lesser degree. First, DaVita plausibly alleges that dialysis is “needed almost exclusively by ESRD patients.” No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 26

Appellant Br. at 41. That is, in a pie chart of dialysis-users, ESRD-diagnosed individuals would take up almost the full pie. See R. 1 (Compl. ¶ 20) (Page ID #7) (“[N]early all enrollees of the Plan who require or will require dialysis are individuals with ESRD who need such treatment to sustain life.”). Moreover, DaVita explains—without objection by the defendants—that individuals with ESRD “need” dialysis with far greater frequency than the rare, non-ESRD users of dialysis do. Compare Appellant Br. at 8 (“[T]he majority of ESRD patients rely on regular dialysis treatments from the time they are diagnosed with ESRD for the rest of their lives.”), with Reply Br. at 14 n.5 (“The few dialysis patients who do not have ESRD typically suffer from acute kidney injury (‘AKI’) and require dialysis for, at most, a period of weeks (as opposed to years).”). The defendants do not counter with their own explanation of what the “need for renal dialysis” means.

The third prohibited basis of anti-ESRD discrimination is the differential treatment of ESRD patients “in any other manner” not covered by the prior two bases. That is, besides engaging in anti-ESRD discrimination by explicitly carving out a different set of benefits for those with ESRD or by targeting these individuals based on their more common and/or frequent need for renal dialysis, a primary plan may not engage in such discrimination through any other means. In DaVita’s view, “[t]he catch-all phrase ‘or in any other manner’ cautions plans against creative devices that have the effect of such illegal differentiation.” Appellant Br. at 42. Again, the defendants offer no competing definition for this type of prohibited discrimination.

Both of these bases support DaVita’s theory of how the Plan discriminated against Patient A in violation of the MSPA and ERISA. As to the second basis, DaVita has plausibly alleged that a principal, distinguishing feature of being diagnosed with ESRD is one’s significant need for renal dialysis. Thus, the Plan discriminates against ESRD patients based on their need for dialysis by targeting the primary treatment that individuals with ESRD (1) need exclusively, with the exception of rare, non-ESRD patients, and (2) need with far greater frequency than those few non-ESRD dialysis-users. MedBen’s primary counterargument to this assertion is that “[t]he Plan unequivocally provides for equal treatment of all dialysis patients in that all outpatient dialysis services for any medical reason are out-of-network.” MedBen Br. at 32. Similarly, before the district court, MedBen focused on the complaint’s reference to “the Plan’s No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 27

general application to ‘dialysis patients, almost all of whom have ESRD,’” implying that DaVita’s failure to allege total overlap between dialysis patients and ESRD patients was fatal. R. 18 (MedBen Mot. to Dismiss at 12) (Page ID #204) (quoting R. 1 (Compl ¶ 55) (Page ID #20)). Yet there are two flaws with this counterargument. First, it addresses only the first basis of unlawful anti-ESRD differentiation set forth in the MSPA, and not the latter two bases. Addressing the lack of explicit discrimination in the Plan based on “the existence of end stage renal disease” is only one-third of the way toward defending the Plan’s legality.

Second, and perhaps more fatally, it represents a flawed understanding of antidiscrimination law. It is true that, by targeting a service rather than a diagnosis, the Plan does not explicitly discriminate against individuals with ESRD. As DaVita notes, however, “[a] direct differentiation (or discrimination) claim does not require that the challenged activity affect only the disfavored group and no one else.” Reply Br. at 15. This is a well-established principle in antidiscrimination jurisprudence. Take the Supreme Court’s decision in Lawrence v. Texas, 539 U.S. 558 (2003), for example. The challenged law criminalized sodomy between two persons of the same sex. Id. at 562. Although same-sex partners are not the only class capable of engaging in sodomy, the Court concluded that the overlap between the two made class-based discrimination apparent: “When homosexual conduct is made criminal by the law of the State, that declaration in and of itself is an invitation to subject homosexual persons to discrimination both in the public and in the private spheres.” Id. at 575. Lawrence also invalidated the Court’s prior holding in Bowers v. Hardwick, 478 U.S. 186 (1986), a decision that had upheld the criminalization of sodomy regardless of whether the participants were of the same sex. See Lawrence, 539 U.S. at 575, 578. This reasoning is common in antidiscrimination jurisprudence. See, e.g., Minneapolis Star & Tribune Co. v. Minn. Comm’r of Revenue, 460 U.S. 575, 581 (1983) (rejecting the argument that Minnesota’s use tax on paper and ink was “part of the general scheme of taxation,” and concluding that the tax singled out the press for differential treatment); Bray v. Alexandria Women’s Health Clinic, 506 U.S. 263, 270 (1993) (“A tax on wearing yarmulkes is a tax on Jews.”). Thus, if discovery in this case reveals that there is, in fact, a “near-perfect overlap between ESRD patients and dialysis patients,” Reply Br. at 14, a jury could No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 28

reasonably conclude that discrimination against the latter constitutes discrimination against the former.

Alternatively, the catch-all provision could support a disparate-impact claim against the Plan. That is, even if the Plan has not directly targeted ESRD patients by differentially treating the service they need far more than anyone else, it may have devised a reimbursement system that has the effect of singling out ESRD patients. The defendants argue that the non- differentiation provision does not permit proof through a disparate-impact theory, but they do so through analyzing the implementing regulations, rather than engaging with the text of the statute and Supreme Court caselaw suggesting that it supports disparate-impact liability.15 In Texas Department of Housing & Community Affairs v. Inclusive Communities Project, Inc., 576 U.S. 519 (2015), the Court considered a provision of the Fair Housing Act (“FHA”) that made it unlawful

[t]o refuse to sell or rent after the making of a bona fide offer, or to refuse to negotiate for the sale or rental of, or otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.

Id. at 533 (quoting 42 U.S.C. § 3604(a)). The case centered on the meaning of “otherwise make unavailable[,]” which the Court determined was “equivalent in function and purpose” to the phrase “otherwise adversely affect” found in both Title VII of the Civil Rights Act (“Title VII”) and the Age Discrimination In Employment Act (“ADEA”), id. at 533–34.16 Because “[i]n these three statutes the operative text looks to results,” the Court held that “otherwise make unavailable”—like the phrase “otherwise adversely affect”—“refers to the consequences of an action rather than the actor’s intent” and “encompasses disparate-impact claims.” Id. at 534.

15Thedistrict court similarly dismissed the disparate-impact argument based on its focus on one subsection of 42 C.F.R. § 411.161(c), without explaining how the MSPA does not “contain th[e] type of ‘results-oriented’ language” of Texas Department of Housing & Community Affairs v. Inclusive Communities Project, Inc., 576 U.S. 519 (2015). See DaVita, 2019 WL 4574500, at *4. 16The Supreme Court held that the phrase “otherwise adversely affect” confers a disparate-impact cause of action in earlier cases. See Griggs v. Duke Power Co., 401 U.S. 424, 426 n.1 (1971) (Title VII); Smith v. City of Jackson, 544 U.S. 228, 234–36 (2005) (ADEA). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 29

As DaVita explains, see Appellant Br. at 47, there are at least two parallels between the provision at issue in Inclusive Communities and the one at issue here. First, the phrase appears at the end of a series of other prohibitions that deal with disparate treatment, counseling in favor of it meaning something different from these other phrases. See Inclusive Communities, 576 U.S. at 534–35 (identifying the “relevant statutory phrases” in the FHA, Title VII, and the ADEA as “[l]ocated at the end of lengthy sentences that begin with prohibitions on disparate treatment”); see also Appellant Br. at 42 (“[F]or these last two phrases to have any meaning, they must be read as expanding the prohibition beyond explicit discrimination against ESRD patients.”). Second, like the phrase “otherwise make unavailable,” the phrase “in any other manner” is exceedingly broad, sweeping in less blatant forms of discrimination. See Inclusive Communities, 576 U.S. at 534–35 (concluding that the “relevant statutory phrases” in the FHA, Title VII, and the ADEA “serve as catchall phrases looking to consequences, not intent”); see also Appellant Br. at 47.

Moreover, the Inclusive Communities Court determined that the word “otherwise” means “‘in a different way or manner,’ thus signaling a shift in emphasis from an actor’s intent to the consequences of his actions.” Inclusive Communities, 576 U.S. at 535 (quoting Webster’s Third New International Dictionary 1598 (1971)) (emphasis added). That “in any other manner” in the MSPA is nearly identical to the Supreme Court’s definition of “otherwise” in Inclusive Communities further reinforces that the non-differentiation provision permits a disparate impact claim.

MedBen’s principal response is that the implementing regulations for the non- differentiation provision foreclose both of DaVita’s theories of discrimination.17 Given the lack of any ambiguity in the statutory text, however, we decline to defer to the implementing

17MedBen also argues that “costs exist under Medicare’s coverage structure for ESRD patients as well,” making it inappropriate “to simply assume . . . that potential exposure to out-of-pocket costs under a private plan serves as an incentive to opt for Medicare.” MedBen Br. at 29. But this argument represents a factual dispute, and at this stage, we assume that all of DaVita’s factual allegations—including DaVita’s allegation that high potential costs for ESRD patients on the Plan incentivize them to switch to Medicare—are true. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 30

regulations. See Perez v. Postal Police Officers Ass’n, 736 F.3d 736, 740 (6th Cir. 2013).18 (“Our analysis begins with the plain meaning and, if the language is unambiguous, ends there as well.”). And even if we viewed the phrases “need for renal dialysis” and/or “in any other manner” as ambiguous terms for which the agency might supply a reasonable interpretation, our analysis of these regulations, as discussed below, points to the same result.

Put simply, the non-differentiation regulations do more to confuse than to clarify, as they appear to conflict with one another. Section 411.161(b)(2) of the Code of Federal Regulations provides examples of unlawful differentiation in plan benefits, including a “[f]ailure to cover routine maintenance dialysis or kidney transplants, when a plan covers other dialysis services or other organ transplants.” 42 C.F.R. § 411.161(b)(2)(v). Routine maintenance dialysis is the exclusive province of ESRD patients; non-ESRD patient uses of dialysis are acute, not routine. See Reply Br. at 20. Thus, limiting benefits for those who rely more frequently on dialysis than others is, per § 411.161(b)(2)(v), unlawful.

Yet the next provision in the regulations seems to suggest the opposite. It reads:

(c) Uniform Limitations on particular services permissible. A plan is not prohibited from limiting covered utilization of a particular service as long as the limitation applies uniformly to all plan enrollees. For instance, if a plan limits its coverage of renal dialysis sessions to 30 per year for all plan enrollees, the plan would not be differentiating in the benefits it provides between plan enrollees who have ESRD and those who do not.

42 C.F.R. § 411.161(c). This provision, unlike its neighbor, appears to let plans create limitations on benefits that target and/or adversely affect those who rely frequently on dialysis. Capping dialysis sessions at thirty per year would impact all plan participants with ESRD but have zero impact for all or almost all individuals without ESRD, who use dialysis sparingly and for short periods of time—a result this provision seems to endorse. No court has considered the internal conflict between these regulatory provisions. Cf. Dialysis of Des Moines, LLC v.

18Nor do we hinge our analysis on the legislative history of the MSPA. See Isle Royale Boaters Ass’n v. Norton, 330 F.3d 777, 784 (6th Cir. 2003) (“When a statute’s text is unambiguous, there is ordinarily no need to review its legislative history.”). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 31

Smithfield Foods Healthcare Plan, No. 2:18-CV-653, 2019 WL 8892581, at *5 (E.D. Va. Aug. 5, 2019) (citing 42 C.F.R. § 411.161(c) without discussion).

Yet even if we considered § 411.161(c) in isolation, we could not rely on a provision that conflicts with clear congressional intent. See Moses v. Providence Hosp. & Med. Ctrs., Inc., 561 F.3d 573, 583 (6th Cir. 2009) (“Because the CMS rule is contrary to the plain language of the statute, this Court does not afford it Chevron deference.”). Congress has forbidden primary plans from treating individuals with ESRD differently based on their “need for renal dialysis,” and this regulatory provision permits treating individuals who need dialysis more often (all or almost all of whom have ESRD) worse than individuals who need it less often (none or almost none of whom have ESRD). This plainly conflicts with the text of the MSPA, not to mention the other provisions of the same implementing regulation, see 42 C.F.R. § 411.161(b)(1), (b)(2)(v).

For these reasons, DaVita has plausibly alleged that the Plan violates the non- differentiation provision of the MSPA, resulting in a denial of benefits and unlawful discrimination under ERISA.

2. The Take-Into-Account Provision

Similar reasoning extends to DaVita’s claim that the Plan violates the take-into-account provision of the MSPA. As a reminder, this provision states that a group health plan “may not take into account that an individual is entitled to or eligible for [Medicare benefits due to ESRD]” during the thirty-month period when the plan is primary to Medicare. 42 U.S.C. § 1395y(b)(1)(C)(i). The statute itself does not define what “take into account” means. DaVita argues that to take something into account means to consider or think of it, which means that plans are prohibited from adopting policies that are motivated by a desire to treat Medicare- entitled individuals differently. See Appellant Br. at 51–52. In DaVita’s view, discovery may reveal evidence of the defendants’ illicit motive. Id. at 52. The defendants do not directly address why DaVita’s motive-based definition of “take into account” is incorrect. Instead, they argue that “take into account” speaks to a plan’s terms, which, in their view, were nondiscriminatory. See MedBen Br. at 39 (“Because all plan beneficiaries who use outpatient dialysis services, regardless of their diagnosis or condition and including those ESRD patients No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 32

who are eligible for Medicare, are subject to out-of-network reimbursement rates, the Plan does not violate the MSPA’s ‘take into account’ prohibition.”). Because “take into account” appears susceptible to these conflicting meanings, it is ambiguous. See Brilliance Audio, Inc. v. Haights Cross Commc’ns, Inc., 474 F.3d 365, 372 (6th Cir. 2007) (“As both parties have laid out plausible readings of the statutory language, we find that the language of § 109(b)(1)(A) is not unambiguous.”). We thus consider whether the agency has construed this term, and if so, whether its construction is permissible. See Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 843 (1984); Bio-Med, 656 F.3d at 282 (deferring to 42 C.F.R. § 411.108’s “reasonable interpretation” of the “take into account” clause).

Section 411.108 of the Code of Federal Regulations sets forth a nonexclusive list of “[e]xamples of actions that constitute ‘taking into account’” that a person is entitled to Medicare on the basis of ESRD, age, or disability. 42 C.F.R. § 411.108(a). At least two of § 411.108’s illustrations buoy DaVita’s motive-based interpretation of the ambiguous statutory phrase: “taking into account” includes “denying or terminating coverage because an individual is entitled to Medicare on the basis of disability without denying or terminating coverage for similarly situated individuals who are not entitled to Medicare on the basis of disability[,]” id. § 411.108(a)(4) or “[i]mposing limitations on benefits for a Medicare entitled individual that do not apply to others enrolled in the plan, such as providing less comprehensive health care coverage, excluding benefits, reducing benefits, charging higher deductibles or coinsurance, providing for lower annual or lifetime benefit limits, or more restrictive pre-existing illness limitations[,]” id. § 411.108(a)(5). These two examples support DaVita’s theory of discrimination for the same reasons discussed above in Part III.C.1.19 DaVita’s central allegation in this case is that ESRD patients, who are, by law, “Medicare[-]entitled individuals,” id., are singled out for differential treatment because their costs are expensive and could be shifted to Medicare. If DaVita shows, through discovery, a “near-perfect overlap” between Medicare-entitled patients (via ESRD diagnosis) and dialysis patients, Reply Br. at 14, then it may show that, compared to other Plan enrollees, Medicare-entitled individuals are subject to

19Our holding does not preclude a disparate impact view of the take-into-account clause. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 33

reduced benefits. For these reasons, DaVita has plausibly alleged a violation of the “take into account” provision of the MSPA for purposes of its ERISA claims.

***

Discovery will permit DaVita the opportunity to demonstrate that the Plan provided Patient A differential benefits based on the Patient A’s “need for renal dialysis” or “in any other manner,” or took into account Patient A’s eligibility for Medicare. If DaVita is successful, it may demonstrate a violation of the MSPA, and in turn, two violations of ERISA. First, as alleged in Count II of the complaint, violating the MSPA’s antidiscrimination provisions would mean that the Plan did not reimburse DaVita “pursuant to the terms of the Plan document and other applicable law.” R. 1 (Compl. ¶ 66) (Page ID #22). Section 1132(a)(1)(B) of ERISA states that in such a circumstance, an entity has a private right of action “to recover benefits due to [it] under the terms of [the] plan.” 29 U.S.C. § 1132(a)(1)(B). Second, as alleged in Count VII of the complaint, discriminating against individuals with ESRD would violate ERISA’s antidiscrimination provision, see 29 U.S.C. § 1182. For the same reasons that DaVita has plausibly alleged unlawful discrimination under the MSPA, as discussed above, it has alleged unlawful discrimination under § 1182. Section 1132(a)(3), in turn, provides a private right of action to enjoin such violations of ERISA. In short, if DaVita is able to prove that the defendants engaged in unlawful discrimination under the MSPA, it would thus demonstrate that, under § 1132(a)(1)(B) and § 1182(a)(1) of ERISA, Patient A was denied benefits due under the Plan and suffered unlawful discrimination, respectively.

IV. CONCLUSION

For the foregoing reasons, we AFFIRM in part, REVERSE in part, and REMAND for discovery and further proceedings on Counts I, II, and VII of DaVita’s complaint. No. 19-4039 Appendix to opinion of Moore, J. Page 34 DaVita, Inc. v. Marietta Mem. Hosp.

APPENDIX _________________

42 U.S.C. § 1395y. Exclusions from coverage and medicare as secondary payer … (b) Medicare as secondary payer … (1) Requirements of group health plans … (C) Individuals with end stage renal disease A group health plan (as defined in subparagraph (A)(v))— (i) may not take into account that an individual is entitled to or eligible for benefits under this subchapter under section 426-1 of this title during the 12-month period which begins with the first month in which the individual becomes entitled to benefits under part A under the provisions of section 426-1 of this title, or, if earlier, the first month in which the individual would have been entitled to benefits under such part under the provisions of section 426-1 of this title if the individual had filed an application for such benefits; and (ii) may not differentiate in the benefits it provides between individuals having end stage renal disease and other individuals covered by such plan on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner; except that clause (ii) shall not prohibit a plan from paying benefits secondary to this subchapter when an individual is entitled to or eligible for benefits under this subchapter under section 426-1 of this title after the end of the 12-month period described in clause (i). Effective for items and services furnished on or after February 1, 1991, and before August 5, 1997, (with respect to periods beginning on or after February 1, 1990), this subparagraph shall be applied by substituting “18-month” for “12-month” each place it appears. Effective for items and services furnished on or after August 5, 1997, (with respect to periods beginning on or after the date that is 18 months prior to August 5, 1997), clauses (i) and (ii) shall be applied by substituting “30-month” for “12-month” each place it appears. … No. 19-4039 Appendix to opinion of Moore, J. Page 35 DaVita, Inc. v. Marietta Mem. Hosp.

(2) Medicare secondary payer (A) In general. Payment under this subchapter may not be made, except as provided in subparagraph (B), with respect to any item or service to the extent that— (i) payment has been made, or can reasonably be expected to be made, with respect to the item or service as required under paragraph (1), or (ii) payment has been made or can reasonably be expected to be made under a workmen’s compensation law or plan of the United States or a State or under an automobile or liability insurance policy or plan (including a self- insured plan) or under no fault insurance. In this subsection, the term “primary plan” means a group health plan or large group health plan, to the extent that clause (i) applies, and a workmen’s compensation law or plan, an automobile or liability insurance policy or plan (including a self-insured plan) or no fault insurance, to the extent that clause (ii) applies. An entity that engages in a business, trade, or profession shall be deemed to have a self-insured plan if it carries its own risk (whether by a failure to obtain insurance, or otherwise) in whole or in part. (B) Conditional payment (i) Authority to make conditional payment The Secretary may make payment under this subchapter with respect to an item or service if a primary plan described in subparagraph (A)(ii) has not made or cannot reasonably be expected to make payment with respect to such item or service promptly (as determined in accordance with regulations). Any such payment by the Secretary shall be conditioned on reimbursement to the appropriate Trust Fund in accordance with the succeeding provisions of this subsection. (ii) Repayment required Subject to paragraph (9), a primary plan, and an entity that receives payment from a primary plan, shall reimburse the appropriate Trust Fund for any payment made by the Secretary under this subchapter with respect to an item or service if it is demonstrated that such primary plan has or had a responsibility to make payment with respect to such item or service. A primary plan’s responsibility for such payment may be demonstrated by a judgment, a payment conditioned upon the recipient’s compromise, waiver, or release (whether or not there is a determination or admission of liability) of payment for items or services included in a claim against the primary plan or the primary plan’s insured, or by other means. If reimbursement is not made to the appropriate Trust Fund before the expiration of the 60-day period that begins on the date notice of, or information related to, a primary plan’s responsibility for such payment or other information is received, the No. 19-4039 Appendix to opinion of Moore, J. Page 36 DaVita, Inc. v. Marietta Mem. Hosp.

Secretary may charge interest (beginning with the date on which the notice or other information is received) on the amount of the reimbursement until reimbursement is made (at a rate determined by the Secretary in accordance with regulations of the Secretary of the Treasury applicable to charges for late payments). (iii) Action by United States In order to recover payment made under this subchapter for an item or service, the United States may bring an action against any or all entities that are or were required or responsible (directly, as an insurer or self-insurer, as a third-party administrator, as an employer that sponsors or contributes to a group health plan, or large group health plan, or otherwise) to make payment with respect to the same item or service (or any portion thereof) under a primary plan. The United States may, in accordance with paragraph (3)(A) collect double damages against any such entity. In addition, the United States may recover under this clause from any entity that has received payment from a primary plan or from the proceeds of a primary plan’s payment to any entity. The United States may not recover from a third-party administrator under this clause in cases where the third-party administrator would not be able to recover the amount at issue from the employer or group health plan and is not employed by or under contract with the employer or group health plan at the time the action for recovery is initiated by the United States or for whom it provides administrative services due to the insolvency or bankruptcy of the employer or plan. An action may not be brought by the United States under this clause with respect to payment owed unless the complaint is filed not later than 3 years after the date of the receipt of notice of a settlement, judgment, award, or other payment made pursuant to paragraph (8) relating to such payment owed. … (3) Enforcement (A) Private cause of action There is established a private cause of action for damages (which shall be in an amount double the amount otherwise provided) in the case of a primary plan which fails to provide for primary payment (or appropriate reimbursement) in accordance with paragraphs (1) and (2)(A).

42 C.F.R. § 411.108 Taking into account entitlement to Medicare (a) Examples of actions that constitute “taking into account”. Actions by GHPs or LGHPs that constitute taking into account that an individual is entitled to Medicare on the basis of ESRD, age, or disability (or eligible on the basis of ESRD) include, but are not limited to, the following: (1) Failure to pay primary benefits as required by subparts F, G, and H of this part 411. No. 19-4039 Appendix to opinion of Moore, J. Page 37 DaVita, Inc. v. Marietta Mem. Hosp.

(2) Offering coverage that is secondary to Medicare to individuals entitled to Medicare. (3) Terminating coverage because the individual has become entitled to Medicare, except as permitted under COBRA continuation coverage provisions (26 U.S.C. 4980B(f)(2)(B)(iv); 29 U.S.C. 1162.(2)(D); and 42 U.S.C. 300bb–2.(2)(D)). (4) In the case of a LGHP, denying or terminating coverage because an individual is entitled to Medicare on the basis of disability without denying or terminating coverage for similarly situated individuals who are not entitled to Medicare on the basis of disability. (5) Imposing limitations on benefits for a Medicare entitled individual that do not apply to others enrolled in the plan, such as providing less comprehensive health care coverage, excluding benefits, reducing benefits, charging higher deductibles or coinsurance, providing for lower annual or lifetime benefit limits, or more restrictive pre-existing illness limitations. (6) Charging a Medicare entitled individual higher premiums. (7) Requiring a Medicare entitled individual to wait longer for coverage to begin. (8) Paying providers and suppliers less for services furnished to a Medicare beneficiary than for the same services furnished to an enrollee who is not entitled to Medicare. (9) Providing misleading or incomplete information that would have the effect of inducing a Medicare entitled individual to reject the employer plan, thereby making Medicare the primary payer. An example of this would be informing the beneficiary of the right to accept or reject the employer plan but failing to inform the individual that, if he or she rejects the plan, the plan will not be permitted to provide or pay for secondary benefits. (10) Including in its health insurance cards, claims forms, or brochures distributed to beneficiaries, providers, and suppliers, instructions to bill Medicare first for services furnished to Medicare beneficiaries without stipulating that such action may be taken only when Medicare is the primary payer. (11) Refusing to enroll an individual for whom Medicare would be secondary payer, when enrollment is available to similarly situated individuals for whom Medicare would not be secondary payer.

42 C.F.R. § 411.161 Prohibition against taking into account Medicare eligibility or entitlement or differentiating benefits. (a) Taking into account— (1) Basic rule. A GHP may not take into account that an individual is eligible for or entitled to Medicare benefits on the basis of ESRD during the coordination period specified in § 411.162 (b) and (c). Examples of actions that constitute taking into account Medicare entitlement are listed in § 411.108(a). … (b) Nondifferentiation. (1) A GHP may not differentiate in the benefits it provides between individuals who have ESRD and others enrolled in the plan, on the basis of the existence of ESRD, or the need for renal dialysis, or in any other manner. No. 19-4039 Appendix to opinion of Moore, J. Page 38 DaVita, Inc. v. Marietta Mem. Hosp.

(2) GHP actions that constitute differentiation in plan benefits (and that may also constitute “taking into account” Medicare eligibility or entitlement) include, but are not limited to the following: (i) Terminating coverage of individuals with ESRD, when there is no basis for such termination unrelated to ESRD (such as failure to pay plan premiums) that would result in termination for individuals who do not have ESRD. (ii) Imposing on persons who have ESRD, but not on others enrolled in the plan, benefit limitations such as less comprehensive health plan coverage, reductions in benefits, exclusions of benefits, a higher deductible or coinsurance, a longer waiting period, a lower annual or lifetime benefit limit, or more restrictive preexisting illness limitations. (iii) Charging individuals with ESRD higher premiums. (iv) Paying providers and suppliers less for services furnished to individuals who have ESRD than for the same services furnished to those who do not have ESRD, such as paying 80 percent of the Medicare rate for renal dialysis on behalf of a plan enrollee who has ESRD and the usual, reasonable and customary charge for renal dialysis on behalf of an enrollee who does not have ESRD. (v) Failure to cover routine maintenance dialysis or kidney transplants, when a plan covers other dialysis services or other organ transplants. (c) Uniform Limitations on particular services permissible. A plan is not prohibited from limiting covered utilization of a particular service as long as the limitation applies uniformly to all plan enrollees. For instance, if a plan limits its coverage of renal dialysis sessions to 30 per year for all plan enrollees, the plan would not be differentiating in the benefits it provides between plan enrollees who have ESRD and those who do not.

42 C.F.R. § 411.165 Basis for conditional Medicare payments. (a) General rule. Except as specified in paragraph (b) of this section, the Medicare intermediary or carrier may make a conditional payment if— (1) The beneficiary, the provider, or the supplier that has accepted assignment files a proper claim under the group health plan and the plan denies the claim in whole or in part; or (2) The beneficiary, because of physical or mental incapacity, fails to file a proper claim. (b) Exception. Medicare does not make conditional primary payments under either of the following circumstances: (1) The claim is denied for one of the following reasons: (i) It is alleged that the group health plan is secondary to Medicare. (ii) The group health plan limits its payments when the individual is entitled to Medicare. (iii) Failure to file a proper claim if that failure is for any reason other than the physical or mental incapacity of the beneficiary. (2) The group health plan fails to furnish information requested by CMS and necessary to determine whether the employer plan is primary to Medicare. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 39

_____________________________________________________________________

CONCURRING IN THE JUDGMENT IN PART AND DISSENTING IN PART _____________________________________________________________________

MURPHY, Circuit Judge, concurring in the judgment in part and dissenting in part. Marietta Memorial Hospital offers its employees a group health plan. Like all group health plans, the “Marietta Plan” divvies up a finite pot of funds across the many healthcare services that employees might need. The plan offers varied reimbursement rates for different providers and services. The largest reimbursements go to “preferred providers” that have agreed to provide services at a discount. But the Marietta Plan lacks a preferred provider for dialysis services. According to DaVita, Inc., and DVA Renal Healthcare, Inc. (collectively, “DaVita”), the plan reimburses dialysis at uniquely low rates. The low rates allegedly have a disparate impact on the plan participants who use dialysis services the most—those with end stage renal disease. That disparate impact, DaVita claims, violates the Medicare Secondary Payer Act and the Employee Retirement Income Security Act (ERISA). Like other courts to consider this theory, the district court rejected it. I would affirm. I agree with my colleagues that DaVita was not assigned the claims it seeks to pursue in Counts III through VI. But I respectfully part ways with them on the other counts (Counts I, II, and VII).

First, DaVita’s Count I does not allege a violation of the Medicare Secondary Payer Act. That Act requires a group health plan to reimburse a provider before Medicare if both programs cover a patient’s services, as will typically be the case for Medicare-eligible individuals with end stage renal disease. To prevent private entities from shirking this primary-payer duty, the Act bars them from enacting plan terms that “take into account” a participant’s Medicare eligibility or that “differentiate” between those with end stage renal disease and others in the covered benefits. 42 U.S.C. § 1395y(b)(1)(C). These limits bar plans from targeting Medicare-eligible participants who have end stage renal disease; they do not bar plans from distinguishing between covered services. The Marietta Plan does not violate the limits because it does not target anyone for different benefits. It offers the same benefits to all participants. To be sure, low reimbursement for dialysis may disparately affect those with end stage renal disease (or at least their providers) No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 40

because they use the services more. But I do not read the statutory text to permit this disparate- impact theory. The theory would also prove unworkable in this context in which many services are reimbursed at many different rates. What if a plan’s rates for dialysis are higher than its rates for some services but lower than its rates for others? Which are the proper “comparators”? The statute gives no guidance on questions like these, and I would not read it to require common-law rate regulation.

Second, DaVita’s Counts II and VII mistakenly rely on two inapplicable ERISA sections. The first section gives plan participants a cause of action to enforce “the terms of the plan.” 29 U.S.C. § 1132(a)(1)(B). But DaVita seeks to invalidate, not enforce, those terms. The second bars a plan from adopting “eligibility” rules that discriminate against individuals based on their health status. Id. § 1182(a)(1). But DaVita challenges benefits rules, not eligibility rules.

I. Medicare Secondary Payer Act

In 1972, Congress provided Medicare benefits to individuals with end stage renal disease. See 42 U.S.C. § 426-1. Many of these individuals also had coverage under private health plans. The question thus arose: Who should pay healthcare providers—a private plan or Medicare? Private parties enacted plan terms that made them secondary payers behind Medicare or that denied coverage to Medicare-eligible individuals. See S. Rep. 97-139, at 469 (1981). To save money, Congress responded by amending the “Medicare Secondary Payer Act.” The change required private plans covering individuals with end stage renal disease to be the primary payers for a defined period (originally 12 months, now 30). Pub. L. No. 97-35, § 2146(a), 95 Stat. 357, 800 (1981). To stop plans from avoiding their primary-payer obligations by ending coverage for these individuals, this law eliminated a tax deduction for plans that differentiated in their benefits between individuals with end stage renal disease and others. Id. § 2146(b), 95 Stat. at 801.

In 1989, Congress restructured the Medicare Secondary Payer Act into the format found today at 42 U.S.C. § 1395y(b). Pub. L. No. 101-239, § 6202(b), 103 Stat. 2106, 2229–32 (1989). Two clauses (which I will call the “take-into-account clause” and the “differentiate clause”) continue to limit a plan’s ability to target individuals with end stage renal disease. They provide: No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 41

A group health plan (as defined in subparagraph (A)(v))—

(i) may not take into account that an individual is entitled to or eligible for benefits under this subchapter under section 426-1 of this title during the [30]-month period which begins with the first month in which the individual becomes entitled to benefits under part A under the provisions of section 426-1 of this title, or, if earlier, the first month in which the individual would have been entitled to benefits under such part under the provisions of section 426-1 of this title if the individual had filed an application for such benefits; and (ii) may not differentiate in the benefits it provides between individuals having end stage renal disease and other individuals covered by such plan on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner[.]

42 U.S.C. § 1395y(b)(1)(C). My colleagues hold that DaVita may enforce these clauses under the Act’s private right of action. Id. § 1395y(b)(3)(A). I would not answer that difficult question. I would instead hold that the two clauses quoted above do not prohibit the Marietta Plan’s terms.

A. The Differentiate Clause

The differentiate clause states that a “group health plan” “may not differentiate in the benefits it provides between individuals having end stage renal disease and other individuals covered by such plan on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner[.]” Id. § 1395y(b)(1)(C)(ii). DaVita argues that the Marietta Plan violates this clause because it treats dialysis services differently from all other covered services. The plan has no preferred dialysis provider and offers low reimbursement rates for dialysis services. Yet DaVita does not allege that the plan differentiates between participants who have end stage renal disease and other individuals. All participants receive the same benefits. So this case asks: Does this clause prohibit a plan that treats all participants the same, but provides worse coverage for services commonly used by those with end stage renal disease?

My answer: No. The clause prohibits plans that offer participants with end stage renal disease different benefits from others. A plan cannot, for example, cover dialysis services for all participants except those with end stage renal disease. Yet a plan that uniformly offers the same benefits to all groups does not violate this clause. That is so even if this neutral plan has a disparate No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 42

impact on those with end stage renal disease because it provides lower reimbursement for services that they use. This reading follows from the relevant text, context, regulations, and precedent.

1. Text

The clause says that a “group health plan” may not “differentiate . . . between” groups of “individuals” “in the benefits it provides” “on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner.” Id. This text’s component parts show that the clause prohibits plans that expressly engage in disparate treatment of individuals with end stage renal disease. It does not bar neutral plans that may have a disparate impact on those individuals.

Group Health Plan. Start with the subject. The clause identifies a “group health plan” as the thing that cannot engage in the differentiation. The act uses the definition of “group health plan” from the Internal Revenue Code. Id. § 1395y(b)(1)(A)(v). The code, in turn, defines group health plan as “a plan . . . of, or contributed to by, an employer . . . to provide health care” to, among others, its employees. 26 U.S.C. § 5000(b)(1). A “plan” is “a formal program for specified benefits.” Random House Dictionary of the English Language 1480 (2d ed. 1987); 42 C.F.R. § 411.21 (defining “plan” as “arrangement”). The clause thus regulates the program, not the entity that picks its terms. That choice suggests that it is the plan terms (the arrangement) that may not engage in the forbidden “differentiation.” Cf. 42 C.F.R. § 411.21 (defining “payer” separately from “plan”).

Differentiate Between. Congress’s verb choice indicates what those plan terms may not do: They may not “differentiate . . . between” two categories. The phrase “differentiate between” means “to establish or create the difference between people or things.” McGraw-Hill’s Dictionary of American Idioms and Phrasal Verbs 151 (2005); Random House, supra, at 552. To fall within this clause, therefore, the terms must create differences between the listed categories.

Individuals. The clause next identifies the categories that the plan terms may not create differences between: “individuals having end stage renal disease” and “other individuals covered by such plan.” The clause thus bars terms that establish differences between two groups of individuals; it does not bar terms that establish differences between services. A plan might create No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 43

service differences if it covers outpatient chemotherapy but not outpatient dialysis. Or it might do so if it requires a $20 copayment for cancer drugs, but a $50 copayment for similarly priced dialysis drugs. If, however, the plan applies these coverage choices to all participants, the plan has not established differences between “individuals.” It has treated all individuals equally.

Benefits. The clause also suggests that it might not bar all differentiation between the two groups, but only a subset of distinctions: those that are “in the benefits it provides” to participants. It thus prohibits a plan from giving individuals with end stage renal disease a different “entitlement to have payment made” for a healthcare service as compared to the entitlement offered to other participants for the same service. 42 U.S.C. § 1395d(a); see Webster’s Third New International Dictionary of the English Language 204 (3d ed. 1986).

In Any Manner. Congress lastly added a phrase to ensure that plan terms would not avoid this prohibited differentiation by drawing clever distinctions. The clause notes that a plan may not differentiate between the two groups of individuals “on the basis of the existence of end stage renal disease, the need for renal dialysis, or in any other manner.” (This list likely contains a typo because it makes no sense to say “on the basis of . . . in any other manner.” A regulation parroting the statute thus adds an extra “or,” noting that a plan may not differentiate between individuals “on the basis of the existence of [end stage renal disease], or the need for renal dialysis, or in any other manner.” 42 C.F.R. § 411.161(b)(1).) The first item in this series shows that a plan may not divide participants into one group of individuals who have “end stage renal disease” and another group who do not, and provide different benefits to the two groups. The second item shows that a plan may not avoid that illegal differentiation by changing the label of the first group from individuals with “end stage renal disease” to individuals who “need renal dialysis.”

The third item (“in any other manner”) bars other “ways” or “methods” that plans might establish differences between individuals who have end stage renal disease and others. See 9 Oxford English Dictionary 324 (2d ed. 1989). The “expansive” use of the word “any” bars any similar differentiation between the two groups. Freeman v. Quicken Loans, Inc., 566 U.S. 624, 635 (2012) (citation omitted). Yet this catchall cannot change the nature of the prohibited action. Just as a ban on “running in any manner” would not prohibit walking, so too the use of “any other No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 44

manner” cannot transform a ban on differentiating between individuals into a ban on differentiating between services. As the Supreme Court has noted when describing the word “any”: “Expansive, yes; transformative, no. It can broaden to the maximum, but never change in the least, the clear meaning of the phrase selected by Congress here.” Id.

Putting these phrases together, I read the clause as barring plan terms that give different benefits to individuals with end stage renal disease, either by name or by definitions that impliedly target that group. The text requires courts to ask: Do a plan’s terms offer different benefits to individuals with end stage renal disease? Or, to put it differently, would an end-stage-renal-disease diagnosis change the benefits that a participant receives? If the answer is “no,” the plan is facially neutral and has not differentiated between individuals in a way that the clause prohibits. In this case, moreover, DaVita makes no claim that the Marietta Plan would flunk this neutrality test.

2. Context

The Medicare Secondary Payer Act’s context confirms that we should interpret the differentiate clause to prohibit plans that engage in express disparate treatment of those with end stage renal disease, not neutral plans that have disparate impacts on them. The Act is not a substantive healthcare law like the Affordable Care Act designed to regulate health-plan benefits. Nor is it an antidiscrimination law like Title VII designed to protect against discrimination. Rather, it is a coordination-of-benefits law designed to dictate “the order of payment” when two programs cover the same service. Blue Cross & Blue Shield of Tex., Inc. v. Shalala, 995 F.2d 70, 73 (5th Cir. 1993); S. Rep. 97-139, at 469–70. A broad reading of the differentiate clause—one that bars even neutral plans with disparate impacts—would transform the Act well beyond this coordination-of-benefits domain. Cf. Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612, 1626–27 (2018). But the clause’s text shows that it lacks the same structure as these other broader laws.

Is the Act a substantive healthcare law? No. A comparison of the Act to other healthcare laws shows that it lacks the provisions that Congress uses when regulating the benefits that health plans must offer. Take the Affordable Care Act. It requires certain health plans to cover a minimum “essential health benefits package.” 42 U.S.C. § 300gg-6(a); id. § 18022(b). Or consider ERISA. It, too, requires group health plans to cover various healthcare services. See, No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 45

e.g., 29 U.S.C. § 1185. It also requires plans to provide a level of “parity” between medical benefits and mental-health benefits. Id. § 1185a. But Congress did not write the differentiate clause in either of these ways. The clause imposes neither a substantive mandate that a plan cover services associated with end stage renal disease nor an equality mandate that the plan treat those services like any other. Indeed, the Act’s allowance for Medicare to make secondary payments if a primary plan does not cover a provider’s full charge signals that it does not compel any minimum reimbursement rates. 42 U.S.C. § 1395y(b)(4). If, however, we read this clause to regulate a plan’s neutral benefits decisions, we would effectively impose the requirement that all plans cover services associated with end stage renal disease on the same terms as other services.

Is the Act an antidiscrimination law? No again. The Act lacks the defining features of the specific antidiscrimination laws that the Supreme Court has read to impose disparate-impact liability. See Tex. Dep’t of Hous. and Cmty. Affairs v. Inclusive Cmtys. Project, Inc., 576 U.S. 519, 530–40 (2015) (Fair Housing Act); Griggs v. Duke Power Co., 401 U.S. 424, 429–32 (1971) (Title VII). To begin with, when interpreting these laws to prohibit neutral practices with disparate impacts, the Supreme Court reasoned that their text tied the legality of a defendant’s action “to the consequences of [the] action[.]” Doe v. BlueCross BlueShield of Tenn., Inc., 926 F.3d 235, 242 (6th Cir. 2019). Title VII, for example, contains two catchall clauses, one of which makes it unlawful for employers “otherwise to discriminate” against an employee and the other of which makes it unlawful for employers to “otherwise adversely affect” the employee. Tex. Dep’t of Hous., 576 U.S. at 531 (quoting 42 U.S.C. § 2000e-2(a)). The Court relied only on the latter clause to find disparate-impact liability. Id. It reasoned that the clause’s “operative text looks to results” by requiring courts to consider how a neutral practice affected an employee. Id. at 534. The Court thus found it proper to rely on the practice’s outcome when deciding on its legality. Id.

This rationale is missing here. See Doe, 926 F.3d at 242. The differentiate clause contains no similar “results-oriented” verb. Tex. Dep’t of Hous., 576 U.S. at 535. Its catchall phrase (“in any other manner”) refers to ways in which plan terms might differentiate between the two groups of individuals. I do not read it as considering the effects of non-differentiating plan terms that treat all individuals equally. Oxford English Dictionary, supra, at 324. At most, this clause is analogous No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 46

to Title VII’s first clause (“otherwise to discriminate”). But the Court has not read that clause as creating disparate-impact liability. See Tex. Dep’t of Hous., 576 U.S. at 530–35.

Next, when finding that the antidiscrimination laws impose disparate-impact liability, the Court relied on their “central purpose”: to “eradicate discriminatory practices within a sector of our Nation’s economy.” Id. at 539. That rationale is also missing here. The Act serves a different function: to protect taxpayers. As many courts have recognized, Congress passed the Act “[t]o ‘curb the rising costs of Medicare[.]’” MSPA Claims 1, LLC v. Kingsway Amigo Ins. Co., 950 F.3d 764, 767 (11th Cir. 2020) (citation omitted); Bio-Med. Applications of Tenn., Inc. v. Cent. States Se. & Sw. Areas Health and Welfare Fund, 656 F.3d 277, 282 (6th Cir. 2011).

“The oddity of applying disparate-impact discrimination in this area points in the same direction.” Doe, 926 F.3d at 242. A disparate-impact framework would require a “wholly unwieldy” analysis. Id. (quoting Alexander v. Choate, 469 U.S. 287, 298 (1985)). A health plan covers services for many different health conditions at varied rates. How should courts decide whether a benefits package uniformly offered to all participants has an illegal disparate impact on individuals with end stage renal disease? Suppose the services associated with that disease are reimbursed at the “median” rate—such that many conditions have better coverage and many have worse. What conditions should we compare end stage renal disease to? Or must end stage renal disease have the absolute worst coverage, such that all other conceivable conditions fare better?

Consider, too, that an administrator’s coverage choices might depend on factors outside its control. Here, for example, DaVita challenges Marietta’s failure to enter into a preferred-provider contract with a dialysis provider. Should it matter whether Marietta made good-faith efforts to do so? Should it matter whether a region’s dialysis market is highly concentrated, such that the rates for dialysis are much higher than the rates for other services? What if a preferred dialysis provider dramatically increases its prices? Would a plan be required to retain the provider on threat of a disparate-impact suit? At day’s end, I see no “objective and workable standard for choosing a reasonable benchmark by which to” decide whether a neutral plan has an illegal disparate impact on those with end stage renal disease. Holder v. Hall, 512 U.S. 874, 881 (1994) (plurality op.). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 47

And the subjectivity of this exercise will inevitably give providers leverage to threaten suit (and double damages, 42 U.S.C. § 1395y(b)(3)(A)) whenever they do not like a neutral plan.

Lastly, apart from disparate-impact claims, the antidiscrimination laws typically otherwise bar a neutral practice only if adopted with an invidious intent to harm a protected group. Ricci v. DeStefano, 557 U.S. 557, 577–78 (2009). And a practice’s disparate impact on that group can be evidence of the intent to harm. Tex. Dep’t of Hous., 576 U.S. at 588 (Alito, J., dissenting). That is what the Supreme Court means when it says that a “tax on wearing yarmulkes is a tax on Jews”: A tax on a practice associated with a group can be evidence of “an intent to disfavor” it. Bray v. Alexandria Women’s Health Clinic, 506 U.S. 263, 270 (1993). Here, though, I do not read the differentiate clause to make intent relevant. It regulates a “group health plan,” not the employer who adopts the plan, and so focuses on the express terms rather than the hidden motives. The catchall “in any other manner” confirms that the clause bars any act that qualifies as the forbidden “differentiation”—whether taken with an invidious or innocuous intent. This fact, too, suggests that the clause does not cover neutral plans. If it did, the clause might not allow the standard antidiscrimination defense that an employer adopted the plan for a legitimate reason unrelated to an intent to harm those with end stage renal disease. And even disparate-impact claims come with the standard “business necessity” defense—another defense that this clause’s text might not include. Tex. Dep’t of Hous., 576 U.S. at 541–42.

In sum, any reading that would reach neutral group health plans would depart from the Act’s context and prove unworkable. So even if I found some ambiguity in the clause’s plain text, I would stick with the reading I have chosen: The clause bars differentiation between individuals, not services. See Util. Air Regulatory Grp. v. EPA, 573 U.S. 302, 321–24 (2014).

3. Regulations

Because “Congress has supplied a clear and unambiguous answer to the interpretive question at hand,” I would not defer to how the relevant administrative agency (the Department of Health and Human Services) approached this question if its view differed from my own. Pereira v. Sessions, 138 S. Ct. 2105, 2113 (2018). Still, the regulation implementing this clause largely comports with my view that it prohibits differentiation between individuals. 42 C.F.R. § 411.161. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 48

A subsection entitled “Uniform Limitations on particular services permissible” states: “A plan is not prohibited from limiting covered utilization of a particular service as long as the limitation applies uniformly to all plan enrollees.” Id. § 411.161(c) (emphasis added). This subsection nowhere suggests that courts must consider whether this limit disparately affects those with end stage renal disease. A hypothetical confirms the point: “For instance, if a plan limits its coverage of renal dialysis sessions to 30 per year for all plan enrollees, the plan would not be differentiating in the benefits it provides between plan enrollees who have [end stage renal disease] and those who do not.” Id. Even though this hypothetical limit will have a disparate impact on end-stage-renal-disease patients, the limit does not violate the differentiate clause because it applies to all plan participants equally.

The regulation also identifies four examples of improper “differentiation” that support my reading that the clause prohibits differentiation between individuals, not services. It notes that a plan engages in improper “differentiation” if it: ends coverage for “individuals” who have end stage renal disease; imposes limits on those “persons” “but not on others enrolled in the plan”; charges the “individuals” higher premiums; or reimburses healthcare providers less for services furnished “to individuals who have” end stage renal disease as compared to the reimbursement for the “same services” furnished to others. 42 C.F.R. § 411.161(b)(i)–(iv) (emphases added). The last example would suggest that it is acceptable to pay different rates for different services, so long as the different providers received those different rates for all participants in the plan.

That said, I agree that this regulation includes one outlier example that seems service- based, not individual-based. The regulation suggests that a plan would violate this clause if it failed “to cover routine maintenance dialysis or kidney transplants, when [it also] cover[ed] other dialysis services or other organ transplants.” Id. § 411.161(b)(v). This service-based example cannot be reconciled with the regulation’s allowance for uniform coverage limits. Id. § 411.161(c). It also departs from the other four individual-based examples of improper differentiation. Id. § 411.161(b)(i)–(iv). All told, then, the regulation generally supports my reading and, to the extent it does not, it would lose any right to deference given its internally contradictory nature. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 49

4. Precedent

As far as I am aware, every district court to consider this question has interpreted this clause as I do. See DaVita, Inc. v. Marietta Mem’l Hosp. Emp. Health Benefit Plan, 2019 WL 4574500, at *4 (S.D. Ohio Sept. 20, 2019); Dialysis of Des Moines, LLC v. Smithfield Foods Healthcare Plan, 2019 WL 8892581, at *5 (E.D. Va. Aug. 5, 2019); DaVita, Inc. v. Amy’s Kitchen, Inc., 379 F. Supp. 3d 960, 973 (N.D. Cal. 2019); Nat’l Renal All., LLC v. Blue Cross & Blue Shield of Ga., Inc., 598 F. Supp. 2d 1344, 1354 (N.D. Ga. 2009). These courts hold that the clause does not bar a plan that offers the same benefits “to all enrollees equally,” even if it offers worse coverage for end-stage-renal-disease services. Marietta, 2019 WL 4574500, at *4. So, for example, a plan that uniformly provides lower reimbursement rates for dialysis services does not differentiate between individuals with end stage renal disease and others. See Amy’s Kitchen, 379 F. Supp. 3d at 973; Nat’l Renal All., 598 F. Supp. 2d at 1354. Likewise, a nine-month cap on dialysis coverage does not violate the clause as “long as it is uniform, meaning that it applies to all plan enrollees regardless of” an end-stage-renal-disease diagnosis. Dialysis of Des Moines, 2019 WL 8892581, at *5. This precedent supports my conclusion: Because the Marietta Plan applies the same uniform benefits to everyone, it does not violate this differentiate clause.

B. The Take-Into-Account Clause

The take-into-account clause states that a “group health plan” “may not take into account that an individual is entitled to or eligible for benefits under” 42 U.S.C. § 426-1 for a discrete period of time. 42 U.S.C. § 1395y(b)(1)(C)(i). DaVita again asserts that this clause does not permit group health plans to offer a neutral benefits package that has a disparate impact on those who are Medicare eligible. At the least, DaVita suggests, the neutral plan could violate this provision if an entity adopted its terms with an intent to target Medicare-eligible individuals. Yet, for the same reasons that I have already discussed, I read this clause to bar only group health plans that contain terms expressly targeting Medicare-eligible individuals who are eligible because of their end stage renal disease. The clause, for example, prohibits a plan from disqualifying an individual from coverage if the individual becomes Medicare eligible. See Bio-Med., 656 F.3d at No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 50

282–83. The clause does not, by contrast, prohibit neutral plans that treat Medicare-eligible individuals the same as everyone else—regardless of any disparate impact or plan-sponsor intent.

First, the text compels this reading. Like the differentiate clause, this clause applies to a “group health plan,” not the employer or payer. So it regulates the “formal program” or “arrangement,” not the motives of the “entities” that adopted it. 42 C.F.R. § 411.21; Random House, supra, at 1480. The phrase “take into account” next shows that the clause bars that arrangement from giving “consideration” to, or making “allowance” for, something. 1 Oxford English Dictionary 86 (2d ed. 1989) (defining “account”); Longman Dictionary of Phrasal Verbs 649 (1983); see Bio-Med., 656 F.3d at 282. The something? Unlike the differentiate clause, this clause shifts the focus from all individuals with end stage renal disease to certain individuals with that disease. It says that, for 30 months, the plan may not consider the fact “that an individual is entitled to or eligible for” Medicare benefits under 42 U.S.C. § 426-1. The clause thus has a narrower scope because Medicare does not cover all individuals with end stage renal disease. It instead starts covering individuals only after they have received dialysis treatment for three months (or have had a kidney transplant). Id. § 426-1(b)(1)–(2); see also 42 C.F.R. § 406.13(e)(2). Thus, this clause prohibits plan terms that consider an individual’s Medicare eligibility under § 426-1, not terms that consider an end-stage-renal-disease diagnosis.

When the clause is read in this way, the Marietta Plan does not violate it. DaVita concedes that the plan terms do not facially target Medicare-eligible individuals. Consider this point from a participant’s perspective. A patient with end stage renal disease must seek dialysis treatments under the Marietta Plan’s rates for the first three months. See 42 U.S.C. § 426-1(b)(1). After that period, the patient will become Medicare eligible. Does that change in status affect this Medicare- eligible participant in any way? No, it will not have any effect on the benefits that a participant receives or the reimbursement rate offered to dialysis providers. I thus do not see how the Marietta Plan itself could be said to take into account a participant’s Medicare eligibility. That participant is treated the same as a participant with end stage renal disease who is not Medicare eligible.

Second, the contextual factors I discussed earlier caution against finding liability for neutral plan terms. If anything, these contextual clues apply with even more force here. To begin with, No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 51

nowhere does the take-into-account clause contain the type of “results-oriented” language that the Supreme Court has required for disparate-impact liability. Tex. Dep’t of Hous., 576 U.S. at 535; see Doe, 926 F.3d at 242. In addition, the Medicare Secondary Payer Act uses the same “take into account” phrase in nearby provisions. The Act indicates that a group health plan may not “take into account” that a plan participant is entitled to participate in the general Medicare program because of age or disability. 42 U.S.C. § 1395y(b)(1)(A)(i)(I), (B)(i). A disparate-impact theory thus would not be cabined to this end-stage-renal-disease provision. It would prohibit group health plans from enacting any neutral term that caused a disparate impact on, for example, individuals over 65 who are Medicare eligible. Many diseases and conditions uniquely impact the elderly, and disparate-impact liability would leave plans with no guidance concerning plan terms for these elder-focused services. Additionally, this disparate-impact theory would render superfluous a nearby provision that bars plans from providing inferior benefits to those over 65 as are available to those under that age. Id. § 1395y(b)(1)(A)(i)(II). That illegal age-based differential treatment would also have an obvious disparate impact on Medicare-eligible individuals over 65. It confirms that the take-into-account clause should not be broadly read to impose disparate-impact liability.

Third, while I again find the language clear, regulations confirm my reading that the clause bars discriminatory plan terms that target Medicare-eligible individuals, not neutral plan terms that apply to all participants. See 42 C.F.R. §§ 411.108, 411.161(a)(1). All of the regulatory examples of improper “taking into account” involve plan terms that target Medicare-eligible individuals. A plan may not terminate “coverage because the individual has become entitled to Medicare,” id. § 411.108(a)(3), impose “limitations on benefits for a Medicare entitled individual that do not apply to others enrolled in the plan,” id. § 411.108(a)(5), or pay “providers and suppliers less for services furnished to a Medicare beneficiary than for the same services furnished to an enrollee who is not entitled to Medicare,” id. § 411.108(a)(8). The examples, by contrast, nowhere suggest that a plan could improperly “take into account” Medicare eligibility by providing the same reimbursement for Medicare-eligible and non-eligible individuals if those neutral rates disparately affected the former group. If anything, the examples suggest that a group health plan may apply neutral rules to all individuals. They note that a plan could neutrally extend to Medicare-eligible No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 52

individuals the requirement that an employee must have worked for a year as long as that requirement applied to all other participants. Id. § 411.108(b)(1).

Fourth, precedent again supports my view. The same district courts discussed above agree that a plan only violates this provision when it “treats those eligible for Medicare differently than those who are not.” Marietta, 2019 WL 4574500, at *3; Dialysis of Des Moines, 2019 WL 8892581, at *5; Amy’s Kitchen, 379 F. Supp. 3d at 972–733; Nat’l Renal All., 598 F. Supp. 2d at 1354. To date, moreover, our court has found a violation of the take-into-account clause only when a health plan expressly eliminated coverage for participants who were eligible for Medicare. Bio-Med., 656 F.3d at 282–83. Yet the Marietta Plan falls on the right side of this divide because, unlike the plan in Bio-Medical, it does not target Medicare-eligible individuals for unique treatment. It thus does not violate this take-into-account clause.

II. ERISA

While the parties’ briefing did not focus on the two ERISA causes of action that DaVita asserts in Counts II and VII, I find it unlikely that DaVita could rely on those provisions here.

A. Section 1132(a)(1)(B)

In Count II, DaVita alleges that it may enforce its alleged violation of the Medicare Secondary Payer Act using ERISA’s private right of action in 29 U.S.C. § 1132(a)(1)(B). Yet this provision allows plan participants to sue to enforce their rights under a plan’s terms. It does not allow them to invalidate those terms—as DaVita seeks to do. Section 1132(a)(1)(B) states that a “civil action may be brought” by a plan “participant or beneficiary” “to recover benefits due to him under the terms of his plan, to enforce his rights under the terms of the plan, or to clarify his rights to future benefits under the terms of the plan.” Id. Each ground for suit requires a plaintiff to seek relief “under the terms of the plan.” Id. (emphasis added). A typical suit under § 1132(a)(1)(B) thus involves a participant challenging a denial of benefits under the plan terms. See, e.g., Univ. Hosps. of Cleveland v. Emerson Elec. Co., 202 F.3d 839, 841, 845–47 (6th Cir. 2000); Wilkins v. Baptist Healthcare Sys., Inc., 150 F.3d 609, 611, 615 (6th Cir. 1998). Nowhere does this text suggest that a participant may sue because “the terms of the plan” violate the law. No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 53

Structural clues point in the same direction. A nearby paragraph—§ 1132(a)(3)—shows that Congress knows how to give parties a cause of action to challenge a plan’s legality. Unlike § 1132(a)(1)(B), this paragraph provides a right to sue to enjoin or redress “any act or practice which violates any provision of this subchapter or the terms of the plan[.]” 29 U.S.C. § 1132(a)(3) (emphasis added). Section 1132(a)(3) thus distinguishes suits to remedy violations of ERISA from suits to remedy violations of the plan. It confirms that § 1132(a)(1)(B), which permits only suits to enforce the plan, does not permit suits to challenge the plan as illegal. Cf. Nat’l Ass’n of Mfrs. v. Dep’t of Def., 138 S. Ct. 617, 631 (2018).

Precedent confirms this reading. See Cigna Corp. v. Amara, 563 U.S. 421, 435–37 (2011). In Cigna, beneficiaries suing under § 1132(a)(1)(B) claimed that plan administrators had violated various ERISA provisions by giving deficient notice about a plan’s changes. Id. at 424. The district court agreed. Id. at 431–32. The court also read § 1132(a)(1)(B) to allow it to reform the plan terms and require benefits consistent with its new terms. Id. at 432–34. The Supreme Court reversed, holding that § 1132(a)(1)(B) does not give courts the power to change a plan’s terms and instead merely allows plaintiffs to enforce those terms as written. Id. at 435–37.

This text and precedent show that DaVita will encounter difficulty if it intends to rely on § 1132(a)(1)(B) to enforce the Medicare Secondary Payer Act. DaVita does not seek to enforce rights under the Marietta Plan’s terms. Instead, DaVita complains about those terms. It attempts to use this provision to strike terms that affect its dialysis services and replace them with better terms. But § 1132(a)(1)(B) does not grant a broad right to litigate whether a plan comports with every law on the books. It permits suits to compel defendants to live up to a plan’s terms. “By arguing that the terms of the Plan do not comply with the law, [DaVita] tacitly concedes that the relief [it] seek[s] exists outside the scope of [the Marietta Plan]. And an action attempting to re-write the terms of a plan is unavailable under § 1132(a)(1)(B).” Soehnlen v. Fleet Owners Ins. Fund, 844 F.3d 576, 583 n.2 (6th Cir. 2016). I also do not see how DaVita could fix this problem by seeking to enforce its alleged violation of the Medicare Secondary Payer Act using the ERISA cause of action in § 1132(a)(3), rather than the one in § 1132(a)(1)(B). Section 1132(a)(3) permits equitable relief to remedy a “practice which violates any provision of this subchapter,” not a practice that violates a different law. 29 U.S.C. § 1132(a)(3). No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 54

B. Section 1182(a)(1)

In Count VII, DaVita relies on this ERISA cause of action in § 1132(a)(3) to remedy an alleged violation of a separate ERISA provision barring certain forms of discrimination (not a violation of the Medicare Secondary Payer Act). See id. § 1182(a)(1). Yet this ERISA section applies to a plan’s rules of eligibility, not to its rules concerning covered benefits.

Section 1182(a)(1) provides: “Subject to paragraph (2), a group health plan . . . may not establish rules for eligibility (including continued eligibility) of any individual to enroll under the terms of the plan based on” a list of “health status-related factors[.]” Id. § 1182(a)(1). The list of factors includes things like “Health Status,” “Medical condition,” “Receipt of health care,” and “Disability.” Id. § 1182(a)(1)(A), (B), (D), (H). This text limits a plan’s ability to adopt “specified criteria” for the “entitlement to be considered or chosen” as a “registered or acknowledged member” of the plan. 5 Oxford English Dictionary 140, 277 (2d ed. 1989) (defining “eligibility” and “enroll”). It does not require a plan to provide any types of benefits to enrolled members.

Section 1182(a)(2) includes disclaimers that confirm this reading. It states that § 1182(a)(1) should not be construed “to require a group health plan . . . to provide particular benefits other than those provided under the terms of such plan[.]” 29 U.S.C. § 1182(a)(2)(A). And it adds that § 1182(a)(1) should not be construed “to prevent such a plan . . . from establishing limitations or restrictions on the amount, level, extent, or nature of the benefits or coverage for similarly situated individuals enrolled in the plan[.]” Id. § 1182(a)(2)(B). These disclaimers remove all doubt that § 1182(a)(1)’s restrictions concern the eligibility for joining (or staying in) the plan, and do not regulate the benefits that the plan provides. The regulation addressing this section supports this reading as it expressly permits uniform limitations on benefits for certain diseases. See 29 C.F.R. § 2590.702(b)(2)(i)(B) & (D) (Example 4).

DaVita’s challenge to the Marietta Plan’s dialysis reimbursement rates thus does not fall within § 1182(a)(1). The patient on whose behalf DaVita sued successfully enrolled in the plan and was never subjected to terms that stripped the patient of eligibility for any of the plan’s covered benefits. Indeed, DaVita’s complaint nowhere challenges any eligibility rules at all, let alone ones that discriminate on the basis of health status. DaVita claims only that the Marietta Plan’s uniform No. 19-4039 DaVita, Inc. et al. v. Marietta Mem. Hosp. et al. Page 55

benefits package exposes end-stage-renal-disease patients to higher costs. That is exactly the type of “limitation[] or restriction[] on the amount, level, extent, or nature of the benefits” that is expressly excluded from this subsection’s scope. 29 U.S.C. § 1182(a)(2)(B).

* * *

For these reasons, I respectfully concur in the judgment in part and dissent in part. I would affirm outright.