Randall Callahan U.S. Department of Health and Human Services

939 F.3d 1251

• Court: Court of Appeals for the Eleventh Circuit
• Decided: September 25, 2019
• Docket: 19-11876
• Status: Published

Opinion

Case: 19-11876 Date Filed: 09/25/2019 Page: 1 of 32

[PUBLISH]

IN THE UNITED STATES COURT OF APPEALS

FOR THE ELEVENTH CIRCUIT ________________________

No. 19-11876 ________________________

D.C. Docket No. 1:19-cv-01783-AT

RANDALL CALLAHAN, KATRYNA GRISSON, CANDICE SEAMAN, MICHAEL WINGATE, EMORY UNIVERSITY, d.b.a. Emory University Hospital, HENRY FORD HEALTH SYSTEM, INDIANA UNIVERSITY HEALTH, OREGON HEALTH & SCIENCE UNIVERSITY, PIEDMONT HEALTHCARE, THE RECTOR AND VISITORS OF THE UNIVERSITY OF VIRGINIA, on behalf of its Medical Center, THE REGENTS OF THE UNIVERSITY OF MICHIGAN, on behalf of its academic medical center, Michigan Medicine, SAINT LUKE'S HOSPITAL OF KANSAS CITY, UNIVERSITY OF IOWA, UNIVERSITY OF KANSAS HOSPITAL AUTHORITY, a body politic and corporate and an independent instrumentality of the State of Kansas, UNIVERSITY OF KENTUCKY, VANDERBILT UNIVERSITY MEDICAL CENTER, VIRGINIA COMMONWEALTH UNIVERSITY HEALTH SYSTEM AUTHORITY, THE WASHINGTON UNIVERSITY, Case: 19-11876 Date Filed: 09/25/2019 Page: 2 of 32

BARNES-JEWISH HOSPITAL,

Plaintiffs - Appellants,

versus

UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES, through ALEX M. AZAR II in his official capacity as Secretary of the United States Department of Health and Human Services, UNITED NETWORK FOR ORGAN SHARING,

Defendants - Appellees,

SUSAN JACKSON, CHARLES BENNETT,

Intervenor Appellees.

________________________

Appeal from the United States District Court for the Northern District of Georgia ________________________

(September 25, 2019)

Before WILSON and NEWSOM, Circuit Judges, and COOGLER,* District Judge.

NEWSOM, Circuit Judge:

The liver is one of the human body’s most vital and versatile organs.

Among its 500-some-odd functions, the liver cleans the blood, regulates amino

acids, produces critical proteins, manages blood clotting, and facilitates digestion.

* Honorable L. Scott Coogler, United States District Judge for the Northern District of Alabama, sitting by designation.

2 Case: 19-11876 Date Filed: 09/25/2019 Page: 3 of 32

But that’s when things go right. Far too often—and due to a variety of causes—

things can go wrong, and when they do modern medicine has to step in. For minor

liver complications, medication and dietary changes will usually do the trick.

When liver failure sets in, though—when things go really wrong—there is often

only one long-term solution: transplant.

This case centers on the high-stakes rules that determine which patients—

among the more than 12,000 currently on the national waiting list—receive the

liver transplants they need. In December 2018, a private nonprofit entity tasked by

the Department of Health and Human Services (HHS) with coordinating the

nation’s organ-transplant system adopted a new policy for allocating donated

livers. This suit followed. Plaintiffs, four liver-transplant candidates and more

than a dozen transplant hospitals, challenged the policy in federal district court on

a variety of grounds and moved for preliminary injunctive relief barring the

policy’s implementation. The district court denied the motion, and plaintiffs filed

an interlocutory appeal.

The central question we face is one of regulatory construction. In particular,

we must determine whether 42 C.F.R. § 121.4(b) required the Secretary of HHS to

take two procedural steps that all agree he did not: (1) referral of the new liver-

allocation policy to an entity called the Advisory Committee on Organ

Transplantation and (2) publication of the new policy in the Federal Register for

3 Case: 19-11876 Date Filed: 09/25/2019 Page: 4 of 32

public comment. We hold that the Secretary was not required to do so, and we

therefore affirm—at least in that regard—the district court’s denial of plaintiffs’

preliminary-injunction motion. Because the district court failed to address two of

plaintiffs’ claims, however, we remand for consideration of them in the first

instance.

I

Before diving into the merits, we first need to canvass the statutory and

regulatory landscape, some factual background, and the case’s procedural posture.

Fair warning: This gets complicated.

A

In the United States, organ transplants are a public-private affair. The

National Organ Transplant Act of 1984 requires HHS to appoint and oversee the

Organ Procurement and Transplant Network (OPTN)—a private nonprofit

responsible for coordinating foundational aspects of the nation’s organ-transplant

system. See 42 U.S.C. § 274. Under the Act, the OPTN must maintain a list of

transplant candidates, implement a system for allocating donated organs, and

ensure the organs’ equitable distribution. See id. § 274(b).

While the Act describes the OPTN’s duties in broad strokes, HHS’s

implementing regulation—the “Final Rule”—covers the nitty-gritty, from the

OPTN’s Board of Directors to its record-maintenance policy. See 42 C.F.R.

4 Case: 19-11876 Date Filed: 09/25/2019 Page: 5 of 32

§§ 121.1–.13. Most importantly for present purposes, the Final Rule prescribes the

procedures that the OPTN must follow when developing new organ-transplant

policies, as well as the circumstances under which—and extent to which—HHS

must review those policies. See id. § 121.4.

We’ll get way down into the regulatory weeds in due time, complete with a

dense block quote of the Final Rule’s pertinent text—but for now it’s enough to

summarize the Rule’s key features. As an initial matter, the Final Rule states that

whenever the OPTN proposes any new policy, its Board of Directors must give

OPTN members and other “interested parties” an opportunity to comment on it,

and the Board must “take [those comments] into account” in developing and

adopting the policy. Id. § 121.4(b)(1). Separately, the Rule requires the OPTN to

provide the Secretary of HHS with two types of proposed policies at least 60 days

prior to their intended implementation: (1) those that the OPTN Board

“recommends to be enforceable”1; and (2) those that relate to “such other matters

as the Secretary directs.” Id. at § 121.4(b)(2). Finally, as part of the same

subsection—and as you’ll see soon enough, this is where the debate hinges—the

1 A bit of additional background: None of the OPTN’s adopted policies are, in and of themselves, legally “enforceable” against members of the transplant community; rather, compliance is strictly voluntary. But the OPTN can recommend to the Secretary that he or she make a policy enforceable. If the Secretary does so, any entity that violates the policy risks an enforcement action to terminate its participation in Medicare or Medicaid. 42 C.F.R. § 121.10(c)(1). So far, that hasn’t been necessary. The OPTN has never asked the Secretary to make one of its organ-allocation policies enforceable; voluntary compliance has been excellent. 5 Case: 19-11876 Date Filed: 09/25/2019 Page: 6 of 32

Final Rule requires the Secretary to refer “significant proposed policies” to the

Advisory Committee on Organ Transplantation and to publish those policies in the

Federal Register for “public comment.” Id.

B

An organization called the United Network for Organ Sharing has served as

the OPTN for the past 35 years. In 2013, United Network approved and

implemented the liver-allocation policy that remains in place today. The current

policy distributes livers based on two geographic criteria: “Regions”—11 groups

of states—and “Donation Service Areas” (DSAs)—58 smaller, geographically

irregular areas (within and among states) that surround the entities that United

Network has tasked with collecting donated organs. 2

In recent years, the use of DSAs has come under fire. Critics of the DSA-

based system contend that because DSAs are neither geographically uniform nor

designed to minimize transit of donated organs, reliance on them can lead to

bizarre allocation results. They argue, for instance, that organs can end up

2 Under the current policy, a donated liver is first matched and offered to patients who are Status 1A or 1B—the most gravely ill—and who reside in the DSA or Region where the liver is acquired. If there is no suitable match, the liver is then offered to patients—again, who reside in the same DSA or Region where the liver is acquired—based on their Model for End-Stage Liver Disease (MELD) score, which rates patients from 6 (least ill) to 40 (most ill). If there are no matching candidates in the DSA or Region with a MELD score of 15 or higher, the liver is then offered to outside candidates.

6 Case: 19-11876 Date Filed: 09/25/2019 Page: 7 of 32

traveling greater distances to less-sick patients. 3 Defenders of the DSA-based

system, by contrast, insist that aligning organ allocation with the organ-

procurement organizations encourages communication between the entities that

collect organs and those that perform transplants.

By 2016, United Network had decided that things needed to change. After

more than a year of exploring alternatives, United Network approved a new liver-

allocation policy in December 2017. That policy—which retained DSAs but

reduced their impact on allocation decisions—was set to take effect in December

2018. In May 2018, however, a group of patients awaiting liver transplants filed a

comment with the Secretary pursuant to 42 C.F.R. § 121.4(d) 4 criticizing any

continued use of DSAs in liver-allocation determinations. Two months later, in

July 2018, the Secretary instructed United Network’s Board to scrap the December

2017 policy and adopt a new one that eliminated the use of Regions and DSAs

altogether.

3 Consider the following example, used as an illustration at oral argument: Under the current, DSA-based policy, if a liver becomes available in Charleston, South Carolina, it would be offered to a moderately ill patient in Memphis, Tennessee (600 miles away) before a critically ill patient in Atlanta, Georgia (266 miles away)—and indeed, would have to be flown directly over Atlanta en route to Memphis. 4 “Any interested individual or entity may submit to the Secretary in writing critical comments related to the manner in which the OPTN is carrying out its duties or Secretarial policies regarding the OPTN.” 42 C.F.R. § 121.4(d). The Secretary must then “seek, as appropriate, the comments of the OPTN on the issues raised in the comments” and must “consider the comments in light of the [Act] and the [Final Rule].” Id.

7 Case: 19-11876 Date Filed: 09/25/2019 Page: 8 of 32

United Network went back to the drawing board, but it faced an extremely

tight timeline. The Secretary’s July 2018 instruction imposed a December 3, 2018

deadline for promulgating the new liver-allocation policy. By September, the

Liver and Intestinal Transplantation Committee—a specialized group within

United Network that makes recommendations to the Board—had homed in on two

alternative, DSA-less approaches for allocating livers: the “Acuity Circles” model

and the “Broader 2-Circle” model.5 When the Committee published its policy

proposal on October 6, it identified the Broader 2-Circle model as the “preferred”

policy, but it sought public comment on both options. At the Committee meeting

on November 2, the Broader-2 Circle model prevailed by a narrow 11-9 vote.

United Network’s Board, however, went the other way. On December 3,

2018—the HHS-imposed deadline—the Board adopted the Acuity Circles model.

This model, the Board found, would result in “lower waitlist mortality rate[s]” and

“more equity in access” for liver-transplant candidates. United Network later set

the policy’s implementation date for April 30, 2019.

5 The Acuity Circles model draws concentric circular boundaries at 150, 250, and 500 nautical miles from the donor hospital. The model then offers the donated liver based on the following hierarchy: (1) Status-1 candidates within the 500-mile circle; (2) candidates with MELD scores of at least 37 within the 150-mile circle, then the 250-mile circle, then the 500-mile circle; (3) candidates with MELD scores between 33 and 36 within the 150-mile circle, then the 250-mile circle, then the 500-mile circle; (4) candidates with MELD scores between 15 and 28 within the 150-mile circle, then the 250-mile circle, then the 500-mile circle. The Broader 2-Circle model uses the same distance-based circles, but places a premium on proximity—it gives lower priority to candidates with greater medical urgency who are farther away from the donor hospital. 8 Case: 19-11876 Date Filed: 09/25/2019 Page: 9 of 32

Up to this point, HHS had remained on the sidelines. Of course, the

Secretary had initiated the process by directing United Network to adopt a new,

DSA-less allocation policy. But once the new policy’s development began, HHS

didn’t actively intervene. As particularly relevant here, consistent with HHS’s

treatment of prior organ-allocation policies, the Secretary didn’t refer the new

policy to the Advisory Committee on Organ Transplantation or publish it in the

Federal Register for public comment.

The new policy’s detractors, however, brought HHS into the mix. Just as

critics of the December 2017 policy had done, a group of hospitals that opposed

the new policy filed a comment with the Secretary asking him to suspend the new

policy’s implementation until something better could be developed. This time,

though, the policy survived the challenge. Acting on the Secretary’s behalf, the

Administrator of HHS’s Health Resources and Services Administration responded

to the comment, announcing that no further action was warranted and that the new

policy would take effect as scheduled.

C

On April 22, 2019—eight days before the new policy’s official

implementation date—a collection of hospitals and individual patients sued HHS

and United Network in the United States District Court for the Northern District of

9 Case: 19-11876 Date Filed: 09/25/2019 Page: 10 of 32

Georgia.6 Plaintiffs’ complaint challenged the new liver-allocation policy on three

grounds: (1) that HHS failed to follow legally required procedures during the

development of the new policy, in violation of the Administrative Procedures Act

(APA); (2) that HHS’s and United Network’s actions were—both substantively

and procedurally—arbitrary, capricious, and otherwise not in accordance with the

law, also in violation of the APA; and (3) that HHS’s and United Network’s

conduct violated the Due Process Clause of the Fifth Amendment. Asserting the

same grounds, plaintiffs also filed a motion for a temporary restraining order,

asking the district court to prevent the new policy’s impending implementation.7

Following HHS’s agreement to delay implementation by two weeks, the

district court received expedited briefing and held a hearing on plaintiffs’ TRO

motion. On May 13, 2018, the district court—in an order that addressed only

plaintiffs’ first claim—denied the motion. The following day, the new liver-

allocation policy went into effect for the first time. Its force, however, was short

lived. Plaintiffs immediately noticed this appeal and sought an injunction pending

6 Shortly after the complaint was filed, two liver-transplant candidates—Susan Jackson and Charles Bennett—sought to intervene as defendants. Holding that Jackson and Bennett (who, in July 2018, had brought suit in the Southern District of New York seeking to invalidate the current liver-allocation policy) had “established a significant interest in the subject matter of this litigation,” the district court granted their motion to intervene under Federal Rule of Civil Procedure 24(b). 7 The district court construed plaintiffs’ TRO motion as also requesting a preliminary injunction, in accordance with Federal Rule of Civil Procedure 65. 10 Case: 19-11876 Date Filed: 09/25/2019 Page: 11 of 32

its disposition, which the district court granted. As a result, HHS reinstated the

prior (and once again current) policy.

So here we are. After years of development, thousands of public comments,

and several revisions, the nation’s policy for allocating donated livers hangs in the

balance.

II

The standard for obtaining preliminary injunctive relief is a familiar one.

Such relief is appropriate if—but only if—the movant shows “(1) substantial

likelihood of success on the merits; (2) irreparable injury will be suffered unless

the injunction issues; (3) the threatened injury to the movant outweighs whatever

damage the proposed injunction may cause the opposing party; and (4) if issued,

the injunction would not be adverse to the public interest.” McDonald's Corp. v.

Robertson, 147 F.3d 1301, 1306 (11th Cir. 1998). Because a preliminary

injunction “is an extraordinary and drastic remedy,” relief may not be granted

“unless the movant clearly established the burden of persuasion as to the four

requisites.” Id. (internal quotation marks and citation omitted). 8

Our task on appeal is to determine whether the district court erred in

concluding that plaintiffs failed to satisfy their burden and were therefore ineligible

8 We review the district court’s ultimate decision to deny plaintiffs’ motion for abuse of discretion, but we examine any constituent legal conclusions de novo. See United States v. Endotec, Inc., 563 F.3d 1187, 1194 (11th Cir. 2009). 11 Case: 19-11876 Date Filed: 09/25/2019 Page: 12 of 32

for preliminary injunctive relief. We begin (and for now find that we can end)

with plaintiffs’ first claim—the only one that the district court addressed. In

particular, the district court concluded that plaintiffs had failed to demonstrate a

substantial likelihood of success on the merits of their contention that HHS

neglected to follow legally required procedures during the new liver-allocation

policy’s development. We hold that the district court was right in so concluding, if

not quite for the right reasons. As for plaintiffs’ second and third claims—

respectively, that defendants’ actions in adopting the new policy were arbitrary and

capricious and deprived plaintiffs of due process—we will remand so that the

district court can consider them in the first instance.

Now, for the promised deep dive into 42 C.F.R. § 121.4(b)—the section of

the Final Rule that we summarized earlier. Section 121.4(b) contains two

subsections; its relevant text is as follows:

(b) The [OPTN] Board of Directors shall:

(1) Provide opportunity for the OPTN membership and other interested parties to comment on proposed policies and shall take into account the comments received in developing and adopting policies for implementation by the OPTN; and

(2) Provide to the Secretary, at least 60 days prior to their proposed implementation, proposed policies it recommends to be enforceable under § 121.10 (including allocation policies). These policies will not be enforceable until approved by the Secretary. The Board of Directors shall also provide to the 12 Case: 19-11876 Date Filed: 09/25/2019 Page: 13 of 32

Secretary, at least 60 days prior to their proposed implementation, proposed policies on such other matters as the Secretary directs. The Secretary will refer significant proposed policies to the Advisory Committee on Organ Transplantation established under § 121.12, and publish them in the Federal Register for public comment. The Secretary also may seek the advice of the Advisory Committee on Organ Transplantation established under § 121.12 on other proposed policies, and publish them in the Federal Register for public comment. . . .

42 C.F.R. § 121.4(b) (emphasis added).

Plaintiffs’ first claim centers on a single sentence of § 121.4(b)(2)—which

we’ve italicized and which, to keep matters straight here, we’ll call the “significant

proposed policies” sentence. Plaintiffs’ contention is simple and straightforward:

Clearly, they say, a policy that fundamentally alters liver-allocation procedures

throughout the country is “significant.” And because it’s undisputed that the

Secretary neither referred the new liver-allocation policy to the Advisory

Committee nor published it in the Federal Register—as, on plaintiffs’ reading, the

significant-proposed-policies sentence requires—it follows that the Secretary

violated § 121.4(b)(2).

Defendants see it differently. They argue that, properly understood,

§ 121.4(b)(2)’s significant-proposed-policies sentence doesn’t apply here at all.

Under defendants’ reading, the significant-proposed-policies sentence’s referral

and publication requirements are triggered only in the two circumstances specified

in § 121.4(b)(2)’s opening clauses: (1) when the policy at issue is one that the 13 Case: 19-11876 Date Filed: 09/25/2019 Page: 14 of 32

OPTN’s Board “recommends to be enforceable”—we’ll call this (more than a little

clunkily) the “recommends to be enforceable” sentence—or (2) when the policy at

issue is one that relates to “such other matters as the Secretary directs”—here, the

“as the Secretary directs” sentence. Because it’s undisputed that neither of those

two conditions obtained here, defendants contend, the Secretary wasn’t required to

refer the policy to the Advisory Committee or publish it in the Federal Register.

Boiled to its bare essence, then, the interpretive question we face is whether

§ 121.4(b)(2)’s referral and publication requirements apply to all “significant

proposed policies” (plaintiffs’ reading) or, instead, only to those “significant

proposed policies” that (1) the OPTN’s Board has “recommend[ed] to be

enforceable” or (2) pertain to a matter that the Secretary has “direct[ed]”

(defendants’ reading). In short: Does § 121.4(b)(2)’s significant-proposed-

policies sentence stand alone, such that it applies universally, or is it modified and

limited by § 121.4(b)(2)’s recommends-to-be-enforceable and as-the-Secretary-

directs sentences?

The district court answered this question in defendants’ favor. Deferring to

HHS’s interpretation of § 121.4(b)(2), that court held that plaintiffs had not

demonstrated a substantial likelihood of success on their claim that the Secretary

was legally obligated to refer the new liver-allocation policy to the Advisory

Committee or publish it in the Federal Register. See Amended Dist. Ct. Order at

14 Case: 19-11876 Date Filed: 09/25/2019 Page: 15 of 32

10–11. We agree with the district court’s bottom-line conclusion—and we

therefore affirm that court’s decision that plaintiffs haven’t shown the requisite

substantial likelihood of success on their first claim—but we arrive at that

conclusion by a different route. As the Supreme Court recently held in Kisor v.

Wilkie, deference “is not the answer to every question of interpreting an agency’s

rules.” 139 S. Ct. 2400, 2414 (2019).9 This case illustrates the truth and wisdom

of that observation. Here, we hold that the “traditional tools of construction”—

which Kisor directs reviewing courts to “exhaust” before resorting to principles of

deference, id. at 2415—provide a clear answer: Defendants’ reading of

§ 121.4(b)(2) is the better one.

We begin, as always—and as Kisor reiterates we should—with “text [and]

structure.” Id.; see also Chase Bank USA, N.A. v. McCoy, 562 U.S. 195, 204

(2011). To orient ourselves, let’s start with the big(ish) picture—a bird’s-eye view.

9 Kisor was issued just before the close of briefing in this case—the day before plaintiffs’ reply brief was due, in fact. In response to this Court’s request, though, the parties filed supplemental briefs addressing the effect, if any, that Kisor should have on this case. We are confident, therefore, that the parties had full opportunity to litigate Kisor’s applicability. We are equally confident that a Kisor-based remand to the district court is unnecessary. Although the district court didn’t have the benefit of the Supreme Court’s decision, Kisor itself disclaims any groundbreaking. See 139 S. Ct. at 2410 (“You might view this Part”—describing the regime of regulatory interpretation and deference associated with Auer v. Robbins, 519 U.S. 452 (1997)— “as ‘just background’ because we have made many of its points in prior decisions.”); see also id. at 2418 (noting that its articulation of Auer restated “longstanding doctrine”).

15 Case: 19-11876 Date Filed: 09/25/2019 Page: 16 of 32

Section 121.4(b)’s two subsections lay out two possible paths for the review and

development of proposed organ-transplant policies. Path number one—

§ 121.4(b)(1)—provides for the usual, baseline OPTN-administered notice-and-

comment review. Subsection (b)(1) places no limitation on the “proposed

policies” to which it applies—it requires OPTN’s Board to give both “OPTN

membership and other interested parties” an opportunity to comment on all

proposed policies, irrespective of substance, and then directs the Board to consider

those comments in developing and adopting the polices for implementation. Path

number two—§ 121.4(b)(2), which governs the Secretary’s involvement in and

review of proposed policies—is different, right off the bat. Unlike subsection

(b)(1), subsection (b)(2) immediately narrows in on two subsets of policies: (1)

those that OPTN’s Board “recommends to be enforceable” and (2) those that relate

to “other matters as the Secretary directs.” Those two types of proposed policies—

and only those—must be provided to the Secretary at least 60 days prior to their

intended implementation. 42 C.F.R. § 121.4(b)(2).

Even before getting into the details—and addressing the significant-

proposed-policies sentence and its placement in subsection (b)(2)—we should

pause to consider what § 121.4(b)’s two-path architecture indicates. It seems, we

think, to imply a “default-and-extra”-style regime, with subsection (b)(1) being the

“default” and subsection (b)(2) being the “extra.” In other words, it suggests—not

16 Case: 19-11876 Date Filed: 09/25/2019 Page: 17 of 32

definitively so, but presumptively—that the baseline (b)(1) requirements apply to

all proposed organ-allocation policies, while the additional (b)(2) requirements

apply only to a subset of those policies.

With that structural context in mind, let’s zoom back in to take a closer look

at § 121.4(b)(2) itself—and in particular, the way in which its several constituent

sentences interact. A careful reading, we think, confirms that defendants’

interpretation is the better one—subsection (b)(2), including the significant-

proposed-policies sentence and its referral and publication requirements, applies

only to two specific types of proposed policies: those that OPTN’s Board

“recommends be enforceable” and those pertaining to matters that the “Secretary

directs.” That is so for at least three reasons.

First, and most fundamentally, there’s the “scope-of-subparts” canon,

pursuant to which, at least as a general proposition, “[m]aterial within an indented

subpart relates only to that subpart.” Antonin Scalia & Bryan A. Garner, Reading

Law: The Interpretation of Legal Texts 156 (2012); see also, e.g., Lary v. Trinity

Physician Fin. & Ins. Servs., 780 F.3d 1101, 1105–06 (11th Cir. 2015)

(“Ordinarily, the scope of a subpart is limited to that subpart . . . .”). Under this

interpretive principle, the significant-proposed-policies sentence’s placement

within subsection (b)(2) indicates that it goes with, relates to, and is limited by the

17 Case: 19-11876 Date Filed: 09/25/2019 Page: 18 of 32

other sentences in that subsection, including the recommends-to-be-enforceable

and as-the-Secretary-directs sentences. 10

Second, we think that is the most natural—and most coherent—reading of

§ 121.4(b)(2). Subsection (b)(2) operates like a flow chart—or, as it was described

at oral argument, a “funnel” that begins wide and narrows as it goes. The first few

sentences mark out the universe of proposed policies to which § 121.4(b)(2)

applies—at the risk of repetition, those that OPTN’s Board “recommends to be

enforceable” and those that relate to “matters [that] the Secretary directs.”

Subsection (b)(2) expressly requires that such proposed polices be provided to the

Secretary for his review 60 days in advance of their implementation. Having

received a proposed policy that fits one of those two descriptions, the Secretary

must then determine whether it constitutes a “significant proposed polic[y].” 42

C.F.R. § 121.4(b)(2). If he concludes that it does, he “will”—must—refer it to the

Advisory Committee and publish it in the Federal Register for public comment. Id.

With respect to “other proposed policies”—i.e., those that he determines are not

“significant”—the Secretary “may,” but need not, refer and publish them. Id. On

defendants’ reading, therefore, the significant-proposed-policies sentence slots in

10 To be sure, where it “makes no sense” to interpret language as being limited to the subpart in which it appears, the scope-of-subparts canon can give way. Lary, 780 F.3d at 1106. For reasons explained in the body, however, that isn’t the case here. Quite the contrary, in fact. 18 Case: 19-11876 Date Filed: 09/25/2019 Page: 19 of 32

comfortably with the sentences that come before and after it—all links in a

continuous chain of administrative review.

Finally, defendants’ reading isn’t just natural, it’s also sensible. As just

explained, § 121.4(b)(2) requires that two particular types of policies be sent to the

Secretary at least 60 days prior to their “proposed implementation”—i.e., before

they are slated to go into full force and effect. Presumably, there’s a reason that

those two—and no others—are singled out that way. If we read the ensuing

significant-proposed-policies sentence as applying—as § 121.4(b)’s text and

structure indicate—only to those two, the reason becomes clear: For “significant

proposed policies” that either OPTN’s Board has “recommend[ed] to be

enforceable” or pertain to a matter that the Secretary has “direct[ed],” the Secretary

must take certain action—in particular, he must refer them to the Advisory

Committee and publish them in the Federal Register. Of course, doing so—all in

advance of the date of “proposed implementation,” id.—requires time. Under

defendants’ reading of § 121.4(b)(2), the recommends-to-be-enforceable and as-

the-Secretary-directs sentences work hand-in-hand with the significant-proposed-

policies sentence: they operate, together, to ensure that the Secretary will have

enough time to do what needs to be done.

If (as plaintiffs insist) HHS had meant for the significant-proposed-policies

sentence to apply beyond the two types of policies flagged at the outset of

19 Case: 19-11876 Date Filed: 09/25/2019 Page: 20 of 32

subsection (b)(2), surely it would have given some textual indication. HHS could,

for instance, have placed the significant-proposed-policies sentence before the

recommends-to-be-enforceable and as-the-Secretary-directs sentences. That might

have signaled that the significant-proposed-policies sentence applied to all

proposed policies, rather than only a subset of them. HHS also could have created

a separate subsection for the significant-proposed-policies sentence. A clear break

from the recommends-to-be-enforceable and as-the-Secretary-directs sentences

would have communicated that the significant-proposed-policies sentence

embodies a distinct, stand-alone requirement. It might (?) even have been possible

for HHS to keep the significant-proposed-policies sentence in its current location

but to say, straight out, that its referral and publication requirements apply to

proposed policies beyond those identified in the recommends-to-be-enforceable

and as-the-Secretary-directs sentences—something like, “The Secretary will

consider all policies proposed by the OPTN and will refer those that he deems

significant . . . .”

But HHS didn’t do any of those things. Instead, it chose to place the

significant-proposed-policies sentence immediately following the recommends-to-

be-enforceable and as-the-Secretary-directs sentences. Given that choice, the

significant-proposed-policies sentence is read most naturally—and in accordance

20 Case: 19-11876 Date Filed: 09/25/2019 Page: 21 of 32

with the scope-of-subparts canon and common sense—as being modified and

limited by those two preceding sentences.

Against all of this, plaintiffs raise two textual arguments—neither of which

convinces us. First, plaintiffs point out that when the significant-proposed-policies

sentence is read alone, it more naturally supports their reading than defendants’.

Maybe so—there’s nothing within the four corners of the significant-proposed-

policies sentence itself that clearly limits its application to the policies referenced

in the recommends-to-be-enforceable and as-the-Secretary-directs sentences. But

as enticingly straightforward as plaintiffs’ argument may be, it’s just not how we

read law—tidbits and fragments in isolation. See, e.g., Strickland v. Water Works

& Sewer Bd. of City of Birmingham, 239 F.3d 1199, 1204–05 (11th Cir. 2001)

(refusing to construe regulatory terms “absent their context”); Scalia & Garner,

supra, at 167 (“Context is a primary determinant of meaning.”). And for reasons

we’ve explained in detail already, when read in toto—and in context—

§ 121.4(b)(2) is best understood as limiting the significant-proposed-policies

sentence’s referral and publication requirements to policies that fit within the

recommends-to-be-enforceable and as-the-Secretary-directs sentences.

Second, plaintiffs invoke the presumption of consistent usage. Under this

canon of construction, a word or phrase is presumed to bear the same meaning

21 Case: 19-11876 Date Filed: 09/25/2019 Page: 22 of 32

throughout a text. See Scalia & Garner, supra, at 170. Plaintiffs assert that

§ 121.4(b)(1) uses the same term that § 121.4(b)(2)’s significant-proposed-policies

sentence uses—“proposed policies”—to mean all proposed policies. According to

plaintiffs, adopting defendants’ interpretation would mean giving the term

“proposed policies” a different, narrower meaning in § 121.4(b)(2)—one that it

bears nowhere else in the text. The response to all of this is basically, “See above.”

Although the presumption of consistent usage has its place, it also “readily yields

to context.” Util. Air Regulatory Grp. v. EPA, 573 U.S. 302, 320 (2014) (internal

quotation marks and citation omitted) (holding that the Clean Air Act’s definition

of “air pollutant” includes greenhouse gases but that the term had a narrower,

context-dependent meaning when it appeared in a section of the statute addressing

permits). And here, read in context, the phrase “proposed policies” in

§ 121.4(b)(2) is most logically limited by the sentences that precede it within the

same subsection.

* * *

Based on the preceding analysis, we hold that defendants’ interpretation of

§ 121.4(b) is demonstrably superior to plaintiffs’. Not only is defendants’ reading

more natural, given the rule’s format, but it also coherently harmonizes the rule’s

several constituent provisions. Plaintiffs’ interpretation, by contrast, wrenches

22 Case: 19-11876 Date Filed: 09/25/2019 Page: 23 of 32

§ 121.4(b)(2)’s significant-proposed-policies sentence out of context and depends

on a wooden literalism that we cannot accept.

Because § 121.4(b)’s text is clear, we needn’t consult extra-textual evidence

concerning “history” and “purpose.” We address these considerations briefly,

however, because they featured so prominently in plaintiffs’ briefing and because,

given the circumstances, we think it prudent to cover the waterfront. See NLRB v.

SW Gen., Inc., 137 S. Ct. 929, 941–42 (2017) (declaring the statutory text “clear,”

but nevertheless considering history- and purpose-based arguments to show why

they were “not compelling”).

The Final Rule (codified in relevant part at 42 C.F.R. § 121.4(b)) was first

published in the Federal Register in 1998 and then amended—to its current

language—in 1999. In conjunction with each of the Rule’s two iterations, HHS

published a statement that purported to describe the agency’s motivations. The

parties cite to these statements extensively, so we consider them in turn.

First, the 1998 Rule. Plaintiffs seize on language from HHS’s 1998

statement, which they insist demonstrates an intent to require broad agency review

of OPTN policies: “The Secretary also recognizes the need for additional public

participation in the development of some OPTN policies, such as fundamental

revisions to organ allocation policies.” 63 Fed. Reg. 16,301. And later: “While

23 Case: 19-11876 Date Filed: 09/25/2019 Page: 24 of 32

we believe that the comment process administered by the OPTN itself is invaluable

in obtaining technical advice, it does not reach all of the affected public . . . or

otherwise provide the functions and protections accorded by the impartial review

by the Secretary.” Id. at 16,310.

But as defendants emphasize, the same HHS statement also appears to give

the Secretary broad discretion to determine how OPTN policies should be

reviewed. Just after “recogniz[ing] the need for additional public participation,”

HHS’s statement observes that the Final Rule accordingly “enable[s] the Secretary

to seek comment from the public and to direct the OPTN to revise policies if

necessary.” Id. at 16,301 (emphasis added). What’s more, the statement goes on

to explain that there is often good reason for the Secretary to employ a more

modest system of review: “A body of voluntary standards that can be rapidly

revised, particularly for purely technical changes, is a crucial function of the OPTN

system and one that the Secretary strongly supports.” Id. at 16,310. If anything,

then, the statement accompanying the 1998 Rule seems to weigh in favor of

defendants’ interpretation, which gives the Secretary the discretion—but does not

impose an obligation—to refer proposed allocation policies to the Advisory

Committee and publish them in the Federal Register.

What, then, of the 1999 amendments? Plaintiffs contend that increased

oversight of OPTN policies was a primary motivation behind the new version of

24 Case: 19-11876 Date Filed: 09/25/2019 Page: 25 of 32

the Final Rule, and to be sure, there’s evidence for that proposition in HHS’s

statement accompanying the amended rule. See, e.g., 64 Fed. Reg. 56,656. The

statement notes, for example, that HHS decided to establish “an independent

scientific review board”—the Advisory Committee on Organ Transplantation—in

order to “help [e]nsure that policies and procedures are evidence-based and guided

by the best available scientific and medical precepts.” Id. at 56,652. And, the

statement continues, HHS amended § 121.4(b)(2) so that the Committee could

“fulfill this . . . responsibility.” Id. According to plaintiffs, these statements prove

that HHS intended the new Advisory Committee to have a broad role in policy

development.

But if you keep reading, HHS’s statement makes it clear that, even with the

1999 revisions, the Secretary’s discretion remains largely intact—the Secretary is

only required to consult the Advisory Committee in a few circumstances. When

HHS’s statement refers to Secretarial review of “recommend[ed] to be

enforceable” policies—which, again, the new liver-allocation policy is not—it uses

mandatory language that mirrors that in § 121.4(b)(2)’s promulgated text: “When

the OPTN proposes enforceable policies, the Secretary will ask the Committee for

its views on the proposals when the proposals are published in the Federal Register

for public comment.” Id. (emphasis added). Contrast that with HHS’s statement

regarding Secretarial review of “other” policies, which, again, tracks the enacted

25 Case: 19-11876 Date Filed: 09/25/2019 Page: 26 of 32

text: “A similar approach may also be used should the Secretary review other

OPTN policies.” Id. (emphasis added). Once again, Secretarial discretion looms

large.

In the end, the regulatory history is—at best—a mixed bag for plaintiffs.

There is, as plaintiffs point out, language in the 1998 and 1999 HHS statements

that speaks generally about the importance of oversight of OPTN policies. But

there is also language consistently recognizing that the Secretary has broad

discretion when it comes to the review of non-“recommend[ed] to be enforceable”

OPTN policies, like the liver-allocation policy at issue here.11

In the end, to the extent they are discernible, § 121.4(b)(2)’s “purpose” and

“history” provide no basis for second-guessing—let alone countermanding—what

its text and structure clearly indicate.

11 To the extent it matters, the post-enactment regulatory history is decidedly not a mixed bag, but rather stands squarely against plaintiffs’ interpretation of § 121.4(b)(2). In the nearly 20 years since the current version of the Final Rule was published, not a single organ-allocation policy has ever been referred to the Advisory Committee or published in the Federal Register. For all the opacity of HHS’s 1998 and 1999 statements, then, its actions since have been clear, and they provide at least some evidence that HHS did not intend to subject non-enforceable organ-allocation policies to § 121.4(b)(2)’s enhanced review procedures. Cf. Kisor, 139 S. Ct. at 2426 (Gorsuch, J., concurring in the judgment) (“[T]he government’s early, longstanding, and consistent interpretation of a statute, regulation, or other legal instrument could count as powerful evidence of its original public meaning.”).

26 Case: 19-11876 Date Filed: 09/25/2019 Page: 27 of 32

There’s one last interpretive issue for us to tackle. Plaintiffs contend that

defendants’ reading of § 121.4(b)(2) violates federal contracting law. In particular,

plaintiffs point to the Federal Activities Inventory Reform (FAIR) Act, Pub. L. No.

105-270, 112 Stat. 2382 (1998), and the Federal Acquisition Regulation, 48 C.F.R.

7.503, which they say together embody a rule that “[c]ontracts shall not be used for

the performance of inherently governmental functions.” Appellants’ Br. at 37–38.

Plaintiffs assert that defendants’ interpretation of § 121.4(b)(2) would “allow[]

[United Network] to perform the inherently governmental function of determining

a policy’s consistency with federal law.” Id. at 38. Plaintiffs’ argument, as we

understand it, is not that United Network violates the FAIR Act and the FAR

regulation simply by developing and proposing new organ-allocation policies, but

rather that if defendants’ interpretation of § 121.4(b)(2) is adopted, United

Network (as the OPTN) would be given a new task—assessing whether its policies

are consistent with federal law—which would violate the FAIR Act and the FAR

regulation.12 We don’t see it that way, for two reasons.

First, it’s not clear why the OPTN would have an affirmative obligation to

ensure its policies’ compliance with federal law under one reading of

12 Although we reject plaintiffs’ FAIR- and FAR-related argument on other grounds, it’s worth noting at the outset that the FAIR Act, in particular, appears to be totally irrelevant to plaintiffs’ claim. The FAIR Act does not, as plaintiffs repeatedly suggest, prohibit the use of contracts for 27 Case: 19-11876 Date Filed: 09/25/2019 Page: 28 of 32

§ 121.4(b)(2), but not another. The specific provision of § 121.4(b)(2) at issue

prescribes the Secretary’s duties, not the OPTN’s. We don’t think that interpreting

the Secretary’s duties in a way that does not require referral of the new liver-

allocation policy to the Advisory Committee or its publication in the Federal

Register would necessarily expand the OPTN’s responsibilities, as plaintiffs’

argument seems to suggest.

Second, even under defendant’s more constrained reading of § 121.4(b)(2),

the OPTN cannot unilaterally evade all Secretarial oversight. The Secretary

maintains significant authority to review OPTN’s proposed policies. As

defendants point out—and as the regulation’s plain text makes clear—the

Secretary can always “direct” OPTN’s Board of Directors to provide him with a

proposed policy 60 days in advance of its implementation, thereby bringing it

within § 121.4(b)(2)’s ambit. Or, wholly separately, under § 121.4(d) the

Secretary can review and suggest revision to any OPTN policy that has been the

subject of a critical comment. Indeed, that’s exactly what happened here; it was

pursuant to the Secretary’s § 121.4(d) authority that HHS decided that the current

the performance of inherently governmental functions. In fact, plaintiffs’ argument seems to rest on an understanding of the FAIR Act that is 180° wrong. The FAIR Act is a reporting statute that requires federal agencies to annually publish lists of activities that are performed by government employees (not private contractors) and are not (rather than are) inherently governmental functions. See 112 Stat. 2382 (1998). 28 Case: 19-11876 Date Filed: 09/25/2019 Page: 29 of 32

liver-allocation policy—the one that plaintiffs favor—“cannot be justified” under

the Final Rule and must be replaced.

Contrary to plaintiffs’ suggestions, therefore, defendants’ interpretation of

§ 121.4(b)(2) does not give the OPTN—here, United Network—free reign over the

country’s organ-allocation policy. The Secretary maintains important—and

significant—oversight over the process, as the facts of this case themselves

demonstrate.

The bottom line: Defendants’ interpretation of § 121.4(b) clearly makes the

most sense of the regulation’s text and structure, and none of the remaining tools of

construction remotely displace that conclusion. We are confident, therefore—even

without resorting to principles of agency deference—that defendants’

interpretation should prevail. Accordingly, we hold that the district court did not

err in concluding that plaintiffs failed to demonstrate a substantial likelihood of

success on the merits of their claim that the Secretary neglected to follow legally

required procedures during the new liver-allocation policy’s development. 13

13 Because we hold that the district court correctly concluded that plaintiffs had failed to demonstrate a substantial likelihood of success, we need not consider whether the remaining factors weigh in favor of a preliminary injunction. See GeorgiaCarry.Org, Inc. v. U.S. Army Corps of Eng’rs, 788 F.3d 1318, 1329 (11th Cir. 2015) (“Because the plaintiffs have not shown a substantial likelihood of success on the merits, we need not consider the remaining factors in the preliminary injunction test.”); ACLU of Fla., Inc. v. Miami-Dade Cty. Sch. Bd., 557 F.3d 1177, 1198 (11th Cir. 2009) (“Failure to show any of the four factors [of the preliminary injunction 29 Case: 19-11876 Date Filed: 09/25/2019 Page: 30 of 32

What of plaintiffs’ remaining claims? In its order denying plaintiffs’ motion

for temporary injunctive relief, the district court didn’t address plaintiffs’

alternative contentions that HHS’s and United Network’s actions in adopting the

new liver-allocation policy (1) were both arbitrary and capricious and (2) violated

the Fifth Amendment’s Due Process Clause. We are wary of diving head-first into

claims that the district court hasn’t yet considered, and we are especially wary of

doing so with respect to these claims—both of which will likely turn on fact- and

context-intensive questions that the district court is better equipped to decide in the

first instance.

Plaintiffs’ arbitrary-and-capricious claim, for instance, depends in part on

the premise that United Network constitutes an “agency” within the meaning of the

APA. 5 U.S.C. § 701(b)(1). And that question—which, so far as we can tell, has

yet to be addressed by any federal court—turns on whether United Network

exercises “substantial independent [government] authority.” Dong v. Smithsonian

Inst., 125 F.3d 877, 881 (D.C. Cir. 1997). Similarly, for plaintiffs to have a

cognizable due process claim against United Network, its actions in adopting the

new policy must be considered “state action”—a question that turns on whether

test] is fatal, and the most common failure is not showing a substantial likelihood of success on the merits.”). 30 Case: 19-11876 Date Filed: 09/25/2019 Page: 31 of 32

United Network’s conduct “resulted from the exercise of a right or privilege

having its source in state authority” and whether United Network can “be

described in all fairness as a state actor.” Edmonson v. Leesville Concrete Co., 500

U.S. 614, 620 (1991) (citation omitted).

And beyond those threshold issues, more fact-dependent questions await.

For plaintiffs’ arbitrary-and-capricious claim: Was HHS’s decision to direct the

new policy’s development based on sufficient evidence? Did United Network use

the appropriate procedures in considering the new policy? Did it adequately

consider public comments? What role, if any, did the expedited timeline play?

Does the new policy substantively comply with statutory and regulatory

requirements? And for plaintiffs’ due process claim: Do plaintiffs have a life,

liberty, or property interest that has been affected by the new policy? If so, did

United Network’s policies adequately afford them an opportunity to be heard? Did

HHS’s?

We think that these questions, which are unavoidably fact-sensitive, should

be addressed first by the district court. See Access Now, Inc. v. Southwest Airlines

Co., 385 F.3d 1324, 1331 (11th Cir. 2004) (noting that this Court is particularly

hesitant to address “fact-bound issues” not considered by the district court). We

are, after all, a court of review, not a court of first view. See, e.g., Bartholomew v.

31 Case: 19-11876 Date Filed: 09/25/2019 Page: 32 of 32

AGL Res., Inc., 361 F.3d 1333, 1341 n.5 (11th Cir. 2004). We therefore remand

plaintiffs’ remaining claims to the district court for its consideration.

III

For the foregoing reasons, we hold that plaintiffs have not shown a

substantial likelihood of success on the merits of their first claim—their allegation

that the Secretary failed to follow legally required procedures under 42 C.F.R.

§ 121.4(b) during the new liver-allocation policy’s development. We also hold,

however, that the district court should decide plaintiffs’ remaining claims—their

APA-based arbitrary-and-capricious claim and their Fifth Amendment Due Process

claim—in the first instance. Accordingly, the district court’s order denying

plaintiffs’ motion for a temporary restraining order is AFFIRMED in part and

REMANDED in part.