UNIVERSITY OF COLORADO HEALTH AT MEMORIAL HOSPITAL et al v. BURWELL

Filing 203

MEMORANDUM OPINION granting in part and denying in part 184 Plaintiffs' Motion to Complete Administrative Records; granting in part and denying in part 185 Plaintiffs' Motion for Summary Judgment, and granting in part and denying in part 188 Defendant's Cross-Motion for Summary Judgment. See document for details. Signed by Judge Rudolph Contreras on 06/17/2022. (lcrc1)

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
UNIVERSITY OF COLORADO HEALTH
AT MEMORIAL HOSPITAL, et al.,
Plaintiffs,
v.
XAVIER BECERRA, Secretary of
Health and Human Services,
Defendant.

:
:
:
:
:
:
:
:
:
:
:

Civil Action No.:

14-1220 (RC)

Re Document Nos.:

184, 185, 188

MEMORANDUM OPINION
GRANTING IN PART AND DENYING IN PART PLAINTIFFS’ MOTION FOR SUMMARY JUDGMENT;
GRANTING IN PART AND DENYING IN PART DEFENDANT’S CROSS- MOTION FOR SUMMARY
JUDGMENT; AND GRANTING IN PART AND DENYING IN PART PLAINTIFFS’ MOTION TO
COMPLETE ADMINISTRATIVE RECORDS
I. INTRODUCTION
In advance of each fiscal year (“FY”), the Secretary of the Department of Health and
Human Services (“HHS” or the “Secretary”) engages in a notice-and-comment rulemaking
process to establish a number that will play a significant part in determining the extent to which
hospitals will receive Medicare reimbursement payments for certain extraordinarily costly
services performed during the fiscal year. Plaintiffs, a group of hospitals, ask the Court to vacate
the rules for FYs 2007–2013 because of alleged procedural and/or substantive defects in HHS’s
rulemaking proceedings for these years. The Court holds that the Secretary’s explanations of
certain decisions reached in the FY 2012 and FY 2013 rules were inadequate under the
Administrative Procedure Act; accordingly, it remands these rules to the Secretary for further

explanation. Otherwise, the Court holds that the Secretary acted lawfully in promulgating the
remaining challenged rules.
II. BACKGROUND
A. Regulatory Framework
The Court assumes familiarity with its detailed descriptions of the regulations governing
the Medicare outlier payments program found in prior opinions in this case. See Mem. Op.
Granting in Part and Denying in Part Def.’s Partial Mot. Dismiss and Granting in Part and
Denying in Part Pls.’ Mot. Suppl. Admin. R. (“Mot. Dismiss Op.”), ECF No. 155; Mem. Op.
Granting Def.’s Mot. Leave to Suppl. Answer (“Mem. Op. Suppl.”), ECF No. 89; Mem. Op.
Granting Def.’s Mot. for Clarification (“Clarification Op.”), ECF No. 57; Mem. Op. Granting in
Part and Denying in Part Pls.’ Mot. to Compel Prod. of Complete Admin. R. (“Suppl. Rec.
Op.”), ECF No. 47. And it will provide additional detail as necessary throughout its analysis.
Still, for orientation, the Court directly repeats, with some modifications, part of the background
it provided in its most recent opinion in this case. Mot. Dismiss Op. at 1–9.
Under Medicare, the federal government reimburses hospitals for supplying medical
services to the elderly and disabled. See Social Security Amendments of 1965 (“Medicare Act”),
Pub. L. No. 89–97, tit. XVIII, 79 Stat. 286, 291.1 Providers are not reimbursed for the full costs
that they incur; instead, they are paid at fixed rates for different categories of services and
treatments, known as “diagnosis-related groups” (“DRGs”). See Billings Clinic v. Azar, 901
F.3d 301, 303 (D.C. Cir. 2018) (citation omitted). However, hospitals are also eligible for
certain outlier payments as a form of protection against unusually complicated and costly cases.
Id. at 304 (citing 42 U.S.C. § 1395ww(d)(5)(A)(ii)). These payments become available when the

1

Codified as amended in 42 U.S.C. § 1395 et seq.

2

provider’s (1) “cost-adjusted charges” for a case exceed (2) the sum of (2a) the default
reimbursement payment and (2b) a fixed dollar amount (known as the “outlier threshold” or the
“fixed loss threshold” (FLT) and determined by the Secretary through an annual rulemaking
process). Id. at 304 (citation omitted).
That first figure—the provider’s “cost-adjusted charges”—is intended to estimate the
provider’s real cost of care, without any markups, and is calculated by multiplying a provider’s
actual charges by a historical “cost-to-charge ratio.” Id. at 304–05 (citation omitted). The
second figure—the sum of the base reimbursement plus the fixed loss threshold—is known as
the “fixed-loss cost threshold.” Id. at 304 (citation omitted). Cost-adjusted charges above the
fixed-loss cost threshold are reimbursed at a rate intended to approximate the marginal cost of
care, currently set at 80 percent in most cases. Id. at 305 (citation omitted).
As an example: imagine a hospital charges $100,000 for an unusually complicated
procedure.2 The $100,000 will be multiplied by a cost-to-charge ratio (“CCR”) (imagine it’s
72:100 or 72 percent, which HHS will have calculated based on historical data), leaving $72,000
of cost-adjusted charges. Imagine too that the standard DRG reimbursement rate for this kind of
procedure is $8,000, and the fixed loss threshold set by the Secretary that year is $11,000. The
hospital will automatically receive the base reimbursement of $8,000. And because the costadjusted charges ($72,000) are greater than the fixed-loss cost threshold ($19,000), the hospital is
also eligible for an outlier payment. That payment will be 80 percent of the difference between
the cost-adjusted charges ($72,000) and the fixed-loss cost threshold ($19,000), or $42,400.

This is based on example offered in the Secretary’s opening motion-to-dismiss brief, see
Def.’s Mem. Supp. Mot. Dismiss at 6, ECF No. 139-1, which is in turn drawn from an August
29, 1997, Federal Register notice: Changes to the Hospital Inpatient Prospective Payment
Systems and Fiscal Year 1998 Rates, 62 Fed. Reg. 45,966, 46,011 (Aug. 29, 1997).
2

3

Notice that when the fixed loss threshold is smaller, it is more likely that a hospital will receive
an outlier payment and that any outlier payment received will be greater.
That leaves an important question: how does the Secretary determine each fiscal year’s
fixed loss threshold? Well, Congress has limited the aggregate amount of Medicare outlier
payments to a narrow range: it “may not be less than 5 percent nor more than 6 percent of the
total payments projected or estimated to be made based on DRG prospective payment rates for
discharges in that year.” 42 U.S.C. § 1395ww(d)(5)(A)(iv). To satisfy this directive, HHS
conducts an annual rulemaking to set the fixed loss threshold at a level that it estimates will
result in total payments within the statutorily-determined range (more on that later). See Billings
Clinic, 901 F.3d at 306–07 (citation omitted). Specifically, since 1989, HHS has attempted to set
an annual threshold that will result in total outlier payments being 5.1 percent of all Medicare
payments. Id. at 307. Crucial to the Secretary’s projections are the providers’ estimated future
cost-to-charge ratios. Id. For instance, if HHS overestimates a future year’s cost-to-charge
ratios (expecting, say, 90 percent when it turns out to be 72 percent), then reimbursable, costadjusted charges will be lower than expected—meaning that HHS may have set the fixed loss
threshold too high and therefore be at risk of undershooting its 5.1 percent payment target.
This is all the more important because, in order to fund outlier payments, the Secretary
withholds the predicted 5.1 percent from all other standard reimbursements. See 42 U.S.C. §
1395ww(d)(3)(B). And the Secretary need not take corrective action when the actual outlier
payments differ from the 5.1 percent target. See Dist. Hosp. Partners L.P. v. Burwell, 786 F.3d
46, 51 (D.C. Cir. 2015) (citing County of Los Angeles v. Shalala, 192 F.3d 1005, 1020 (D.C. Cir.
1999)). As a result, undershooting the 5.1 percent target results in a net loss of payments to
providers as a whole.

4

Procedurally, healthcare providers are reimbursed on a rolling basis, but at the end of
their fiscal years, they submit annual cost reports to so-called “medicare administrative
contractors” or “fiscal intermediaries.”3 See 42 U.S.C. § 1395h(a); 42 U.S.C. § 1395kk-1(a)(3)–
(a)(4)(B); 42 C.F.R. § 413.20(b). Fiscal intermediaries then issue a total reimbursement
determination for the entire year4 through a Notice of Program Reimbursement (“NPR”). 42
C.F.R. § 405.1803(a). Hospitals are permitted to challenge an NPR by appealing to the Provider
Reimbursement Review Board (“PRRB”), a specialized administrative body. 42 U.S.C. §
1395oo(a). Hospitals can in turn seek judicial review of a PRRB’s final decision. §
1395oo(f)(1). Providers also “have the right to obtain judicial review of any action of the fiscal
intermediary which involves a question of law or regulations relevant to the matters in
controversy whenever the [PRRB] determines . . . that it is without authority to decide the
question”; such determinations for expedited review can be made sua sponte by the PRRB or at
the request of a provider. Id. In either case, a district court reviews the challenged action
“pursuant to the applicable provisions” of the Administrative Procedure Act (“APA”). Id.
One other feature of the process bears mentioning at this stage. In the early 2000s, the
Secretary would determine a hospital’s cost-to-charge ratio using “cost and charge data from the
‘latest available settled cost report’ without any forward projections.” Billings Clinic, 901 F.3d
at 305. But this approach proved problematic, because
cost reports take several years to settle. And that time lag generated opportunities
for abuse. Hospitals could manipulate their outlier payments by inflating current
“Medicare administrative contractor” is the current statutory terminology. See 42
U.S.C. § 1395h(a). Fiscal intermediary is an older term, see Palisades Gen. Hosp. Inc. v. Leavitt,
426 F.3d 400, 401 (D.C. Cir. 2005), but it remains in usage, see, e.g., 42 U.S.C. §
1395oo(a)(1)(A)(i). The Court will use both terms interchangeably to refer to the kind of entities
described in 42 U.S.C. § 1395h(a).
3

Note that a hospital’s fiscal year may not align with the federal fiscal year, meaning that
a single NPR may be governed by two different fiscal year thresholds.
4

5

charges so that the historic cost-to-charge ratio employed to calculate outlier
payments did not reflect the hospital’s true costs. In those situations, the hospital’s
cost-to-charge ratio would overstate actual costs, resulting in an inflated cost
estimate for the current year’s claims.
Id. This trick came to be known as “turbo-charging.” Id. at 306. The Secretary responded to the
turbo-charging problem in 2003 by enacting a series of reforms, including “reserv[ing] the right
to recalculate a hospital’s eligibility [for an outlier payment] using actual cost data at the time of
settlement. Through this process, known as reconciliation, the agency [can] claw-back undue
outlier payments.” Id.(citations omitted).
B. Procedural History
Many of the plaintiff hospitals here were plaintiffs in two other related cases. Banner
Health v. Azar, No. 10-cv-1638 (D.D.C.) was filed in 2010. In addition to advancing some other
claims, the Banner Health plaintiffs challenged the fixed loss threshold determinations for
federal fiscal years 1997 through 2007. Banner Health v. Burwell, 126 F. Supp. 3d 28, 43
(D.D.C. 2015). The district court disposed of the plaintiffs’ claims through various motions to
dismiss and for summary judgment. See Banner Health v. Burwell, 174 F. Supp. 3d 206, 207–08
(D.D.C. 2016). The Circuit largely affirmed, though it reversed the district court’s grant of
summary judgment as to fiscal years 2004 through 2006 on the grounds that HHS inadequately
explained certain aspects of those threshold calculations. See Banner Health v. Price, 867 F.3d
1323, 1337–39 (D.C. Cir. 2017). The parties stipulated to dismissal of that case with prejudice
in June 2020. Order, Banner Health v. Azar, No. 10-cv-1638 (D.D.C. June 18, 2020).
Another group of cases were filed in 2013 and 2014 and were consolidated in Lee
Memorial Hospital v. Burwell, No. 13-cv-643 (D.D.C.). The Lee Memorial plaintiffs challenged
certain rulemaking actions taken in 2003 and the fixed loss threshold determinations for federal
fiscal years 2008 through 2011. The court granted summary judgment for the Secretary on all

6

the plaintiffs’ claims. Lee Mem’l Health Sys. v. Burwell, 206 F. Supp. 3d 307, 336 (D.D.C.
2016). On appeal of those cases (under the caption Billings Clinic v. Azar), the Circuit affirmed,
finding that the calculations were reasonable and that the challenge to the 2003 rulemaking
actions was precluded by Banner Health. See Billings Clinic, 901 F.3d at 302–03.
This case, University of Colorado Health at Memorial Hospital v. Azar, No. 14-cv-1220,
was consolidated for all purposes with seven later-filed cases. See Dec. 19, 2018 Order at 1,
ECF No. 108; Feb. 15, 2019 Order at 1, ECF No. 112; April 1, 2019 Order at 2, ECF No. 131.
The consolidated action comprises eight currently operative complaints from hospitals who
administratively appealed their cost reports from various years on the ground that their payments
on outlier claims were too low. See Pls.’ Mot. Summ. J. & Mem. Supp. at 10 (“Pls.’ Summ. J.
Mem.”), ECF No. 185. Each plaintiff hospital received from the PRRB a grant of expedited
judicial review regarding the validity of the fixed loss threshold regulations governing their cost
reports. See id. This Court dismissed certain claims as voluntarily abandoned and others as
barred by claim preclusion. See id.; Mot. Dismiss Op. at 14, 19.
C. The Instant Motions
The parties have filed cross motions for summary judgment in their favor on all
remaining claims. Pls.’ Summ. J. Mem.; Def.’ Cross Mot. Summ. J., ECF No. 188. In general,
Plaintiffs’ remaining claims assert that HHS failed to follow proper procedures and/or acted
unreasonably when it calculated fixed loss thresholds in rulemakings for fiscal years 2007–2016.
These errors allegedly caused HHS to set the fixed loss threshold too high for these years, which
in turn caused Plaintiffs to receive less in outlier payments than they otherwise would have.
Recall that each year, HHS “must estimate the number of outlier cases for the upcoming
year and set a threshold that it believes will result in outlier payments of 5.1%” of total Medicare
payments for the year. Billings Clinic, 901 F.3d at 307. To accomplish this, HHS “forecast[s]
7

the total outlier payments that it would make under various potential thresholds, compared to
total DRG payments it would make during the upcoming [fiscal year], until it finds a threshold
projected to produce total outlier payments at its 5.1% target.” Pls.’ Summ. J. Mem. at 5.
Generally, during the years at issue, HHS used the following method:
1. Assume patient cases in the coming FY will be the same as those in the MedPAR5
from two FYs before (e.g., for FY 2008, HHS used FY 2006 cases).
2. Simulate DRG-related payments on those cases using payment rates and policies
for the coming FY.
3. Simulate any outlier payments on those cases, using the rules governing outlier
payments and the following process:
a. HHS increased the charges from the MedPAR file by two years’ worth of
charge inflation, assuming charges would increase the same amount
annually as the year-on-year increase reflected in that file.
b. HHS then multiplied these forecasted charges by forecasted CCRs. To
derive forecasted CCRs, HHS took each hospital’s most recently available
CCR in the March [Provider Specific File (“PSF”)]. HHS also applied an
“adjustment factor” to the CCRs, which was necessary because (1) the
CCRs in the PSF would later be updated at least once, and in many cases
twice, before their use to calculate individual outlier claims during the
upcoming FY and (2) HHS had long recognized that CCRs have generally
been decreasing over time. Thus, if CMS did not apply an accurate
adjustment to the CCRs, it would systematically over-project hospital costs
and outlier payments and set excessive thresholds.
Id. at 5–6 (citations and footnotes omitted); see also Billings Clinic, 901 F.3d at 307–08. HHS
continued to apply this method across the FY 2007–2013 threshold-setting rules. See Pls.’
Summ. J. Mem. at 20 n.13 (collecting sources).
Plaintiffs’ claims in this case relate to HHS’s calculation of the adjustment factor applied
to estimate CCRs for the coming year. The FY 2007 rule was the first in which HHS applied an
adjustment factor to forecast CCRs—in previous years, it had simply (at a general level) “take[n]

“MedPAR is a database that aggregates the claims submitted by hospitals to HHS.”
Banner Health v. Price, 867 F.3d 1323, 1342 (D.C. Cir. 2017).
5

8

the historical cost-to-charge ratio from the most recent year available and project[ed] those
figures forward.” Billings Clinic, 901 F.3d at 307. The adjustment factor used in the FY 2007
rule consisted of two parts, one for each component of the cost-to-charge ratio: an estimate of
inflation of charges and an estimate of inflation of costs. The “charge inflation factor” consisted
simply of the average annual rate of change in charges per case over the past two years. Id. at
308; 71 Fed. Reg. 47,870,48,149 (Aug. 18, 2006). The “cost inflation factor” was “more
complex,” and “factored in both hospital-specific cost inflation and general inflation as measured
by the change in a standard market basket of goods and services.” Billings Clinic, 901 F.3d at
308. Thus, “HHS calculated . . . the average increase (over all relevant hospitals) in hospital
operating costs; and it divided that figure by the increase in the ‘market basket’ . . . for the same
time period.” Pls.’ Summ. J. Mem. at 16 (citing 71 Fed. Reg. at 48,150). “HHS then averaged
that number over three prior years.” Id.; see Billings Clinic, 901 F.3d at 308; 71 Fed. Reg. at
48,150.
Plaintiffs claim that this method of projecting cost inflation is both procedurally and
substantively unsound. They say that the Secretary violated the Medicare Act’s procedural
notice and comment requirements, see 42 U.S.C. § 1395hh(b)(1), by failing to announce the new
complex CCR adjustment method in the proposed rule for FY 2007 before implementing it in the
final FY 2007 rule and by failing to publicize certain documents related to the method during the
2007–2013 rulemakings. Pls.’ Summ. J. Mem. at 15–18, 20–21. Plaintiffs also claim that the
FY 2008 rule failed to comply with notice and comment procedures because the final rule
forecasted an increase in CCRs even though the proposed rule estimated a decrease in CCRs. Id.
at 18–19. As for substance, Plaintiffs assert that the Secretary was arbitrary and capricious in
setting the thresholds in various fiscal years because its selection of a three-year averaging period

9

for the cost inflation factor was arbitrary, because it persisted in using the same method even
though its selected outlier thresholds frequently failed to hit the intended 5.1 percent target,
because it did not sufficiently engage with commenters’ alternative suggestion of adjusting its
projected payments to account for its historical rate of underestimation, and because it did not
account in its projections for the possibility that it would recoup outlier payments during the
reconciliation process. Id. at 22, 27, 32–33.
In his cross motion for summary judgment and opposition to Plaintiffs’ motion, the
Secretary contends that each of Plaintiffs’ complaints about the threshold-setting rules is legally
deficient. Mem. Supp. Sec’y’s Cross Mot. Summ. J. & Opp’n Pls.’ Mot. Summ. J. at (“Def.’s
Summ. J. Mem.”), ECF No. 188-1 at 12–30. He also says certain hospitals’ claims are untimely
and/or barred by issue preclusion. Id. at 40–43.
On the same day Plaintiffs moved for summary judgment, they also moved to complete
the administrative records for the FY 2008–2013 rules with two documents purportedly showing
analysis relevant to HHS’s method for calculating CCR adjustment factors. Pls.’ Mot. Complete
Admin. R. at 1, ECF No. 184. Defendant opposes this motion. Sec’y’s Opp’n Pls.’ Mot.
Contesting Admin. R. at 1 (“Def.’s Admin. R. Opp’n”), ECF No. 186.
III. LEGAL STANDARDS
In a typical case, the Court must grant summary judgment to a movant who “shows that
there is no genuine dispute as to any material fact and the movant is entitled to judgment as a
matter of law.” Fed. R. Civ. P. 56(a); see also Winston & Strawn, LLP v. McLean, 843 F.3d 503,
505 (D.C. Cir. 2016). But in the context of the APA, the Court’s review of the administrative
record is limited. Sierra Club v. Mainella, 459 F. Supp. 2d 76, 89 (D.D.C. 2006) (citing Nat’l
Wilderness Inst. v. U.S. Army Corps of Eng’rs, 2005 WL 691775, *7 (D.D.C. Mar. 23, 2005)). It

10

is the agency’s role to resolve issues of fact and regulate in accordance with those facts. See
Sierra Club, 459 F. Supp. 2d at 90.The district court’s review is confined to determining
whether, as a matter of law, the evidence in the administrative record supports the agency’s
decision. Citizens for Resp. & Ethics in Washington v. SEC, 916 F. Supp. 2d 141, 145 (D.D.C.
2013). “Summary judgment thus serves as the mechanism for deciding, as a matter of law,
whether the agency action is supported by the administrative record and otherwise consistent
with the APA standard of review.” Id. (citing Richards v. INS, 554 F.2d 1173, 1177 & n.28
(D.C. Cir. 1977)).
As the foregoing standard suggests, in an APA case, a reviewing court normally “should
have before it neither more nor less information than did the agency when it made its decision.”
IMS, P.C. v. Alvarez, 129 F.3d 618, 623 (D.C. Cir. 1997) (quoting Walter O. Boswell Mem’l
Hosp. v. Heckler, 749 F.2d 788, 792 (D.C. Cir. 1984)). Agencies bear the responsibility of
compiling the administrative record, which must include all of the information that the agency
considered “either directly or indirectly.” Marcum v. Salazar, 751 F. Supp. 2d 74, 78 (D.D.C.
2010). The record that an agency produces “is entitled to a strong presumption of regularity.”
Id.
A party may seek to supplement the record produced by the agency, however, in “one of
two ways.” WildEarth Guardians v. Salazar, 670 F. Supp. 2d 1, 5 n.4 (D.D.C. 2009). First, a
party may seek to include “evidence that should have been properly a part of the administrative
record but was excluded by the agency.” Id. Where a plaintiff follows this first route,
supplementation is appropriate if the agency “did not include materials that were part of its
record, whether by design or accident . . . .” Marcum, 751 F. Supp. 2d at 78. But to overcome

11

the presumption of regularity, “a plaintiff must put forth concrete evidence that the documents it
seeks to ‘add’ to the record were actually before the decisionmakers.” Id.
Alternatively, a party may seek to supplement the record with “extra-judicial evidence
that was not initially before the agency but [which] the party believes should nonetheless be
included in the administrative record.” WildEarth Guardians, 670 F. Supp. 2d at 5 n.4. In these
circumstances, a more stringent standard applies. To “justify[ ] a departure from [the] general
rule” that review “is to be based on the full administrative record that was before the Secretary at
the time he made his decision,” a party must demonstrate one of three “unusual circumstances.”
Am. Wildlands v. Kempthorne, 530 F.3d 991, 1002 (D.C. Cir. 2008) (internal quotation marks
omitted). Those circumstances include: (1) when “the agency ‘deliberately or negligently
excluded documents that may have been adverse to its decision,’” (2) when “background
information [is] needed ‘to determine whether the agency considered all the relevant factors,’”
and (3) when “the ‘agency failed to explain administrative action so as to frustrate judicial
review.’” City of Dania Beach v. F.A.A., 628 F.3d 581, 590 (D.C. Cir. 2010) (quoting Am.
Wildlands, 530 F.3d at 1002).
IV. ANALYSIS: PROCEDURAL CHALLENGES
“Pursuant to the Medicare Act, 42 U.S.C. § 1395oo(f)(1), this Court reviews the
Secretary’s action under the familiar provisions of the APA, 5 U.S.C. § 706(2)(A).” Shands
Jacksonville Med. Ctr. v. Burwell, 139 F. Supp. 3d 240, 261 (D.D.C. 2015). Thus, HHS “must []
provide the public with a meaningful opportunity to comment on a proposed rule and must offer
reasoned responses to significant comments.” Id. Additionally, “[a]n agency may promulgate a
rule that differs from a proposed rule only if the final rule is a ‘logical outgrowth’ of the
proposed rule. A final rule is a logical outgrowth if affected parties should have anticipated that

12

the relevant modification was possible.” Allina Health Servs. v. Sebelius, 746 F.3d 1102, 1107
(D.C. Cir. 2014) (citations omitted); see Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158,
174 (2007) (noting that the logical outgrowth doctrine is an interpretation of the APA’s
requirement that a notice of proposed rulemaking contain “either the terms or substance of the
proposed rule or a description of the subjects and issues involved.” (quoting 5 U.S.C.
§ 553(b)(3))).
A. FY 2007 Rule
Plaintiffs raise two distinct yet related notice-and-comment arguments about the FY 2007
rule—they argue that HHS was required to present its CCR adjustment factor method for
comment before adopting that method in the FY 2007 final rule, and that HHS “withheld a
critical analysis that it had conducted to develop the method” in violation of the D.C. Circuit’s
Portland Cement doctrine, also known as the critical material doctrine. Pls.’ Summ. J. Mem. at
13, 16, 18 (citing the D.C. Circuit’s quotation in Am. Radio Relay League, Inc. v. F.C.C., 524
F.3d 227, 237 (D.C. Cir. 2008) of the standard articulated in Portland Cement Ass’n v.
Ruckelshaus, 486 F.2d 375, 393 (D.C. Cir. 1973), that “[i]t is not consonant with the purpose of a
rule-making proceeding to promulgate rules on the basis of . . . data that, [to a] critical degree, is
known only to the agency”). Neither objection persuades the Court.
1. The final FY 2007 rule was a logical outgrowth of the proposed FY 2007 rule.
In order to evaluate Plaintiffs’ logical outgrowth challenge, it is necessary to maintain a
bit of perspective. This Court previously held that the rules Plaintiffs challenge are “the overall
calculation of a given year’s fixed los[s] threshold”; that is, Plaintiffs did not preserve any
challenge “to the cost-to-charge ratio methodologies as . . . standalone” rules requiring their own
notice-and-comment proceedings. Mot. Dismiss Op. at 25. Therefore, the question is whether
the overall calculation of the fixed loss threshold implemented in the final FY 2007 Rule was a
13

logical outgrowth of the overall calculation of the fixed loss threshold proposed in the
Secretary’s notice of proposed rulemaking for FY 2007. The method of projecting CCRs is just
one element of this overall calculation, albeit a “crucial” one. Id.
In the notice of proposed rulemaking, the Secretary proposed using the same recenthistory-based method for projecting CCRs it had used up to that point:
As we have done in the past, we are proposing to establish the proposed FY 2007
outlier threshold using hospital cost-to charge ratios from the December 2005
update to the Provider-Specific File—the most recent available at the time of this
proposed rule. This file includes cost-to-charge ratios that reflect implementation
of the changes to the policy for determining the applicable cost-to charge ratios that
became effective August 8, 2003 (68 FR 34494). Using this methodology, we are
proposing to establish an outlier fixed-loss cost threshold for FY 2007 equal to the
prospective payment rate for the DRG, plus any IME and DSH payments, and any
addon payments for new technology, plus $25,530.
71 Fed. Reg. 23,996, 24,150 (Apr. 25, 2006); Administrative Record (“AR”) (FY 2007) at
012369. During the comment period, commenters expressed concern that the proposed threshold
was too high and would cause HHS to undershoot its 5.1 percent target. They “recommended
[that HHS] further refin[e] the outlier methodology,” including by “us[ing] an adjustment factor
to project CCRs” in addition to inflating charges. 71 Fed. Reg. at 48,149. “The commenters
believed that the use of more than one indicator [would] make the threshold calculation more
reliable and accurate.” Id. Thus, these commenters “calculated a cost inflation factor of 5.69
percent by determining the 2002-2004 aggregate annual rate of increase in cost per discharge.”
Id. Another commenter suggested using a “cost inflation factor using the market basket when
projecting CCRs.” Id. at 48,150.
After taking these comments into consideration, HHS agreed in the final rule that “a
refinement to the proposed methodology to account for the rate of change in the relationship
between costs and charges would likely increase the precision of [its] model . . . .” Id. Thus, it
implemented the same fixed loss threshold calculation methodology it had proposed in its notice,
14

except that it used more recent data for the charge inflation factor and that, as commenters
suggested, it “appl[ied] an adjustment factor to the CCRs to account for cost and charge
inflation.” Id. Rather than adopting the precise method proposed by any commenter, HHS
“worked with [its] actual office in deriving” the cost adjustment factor methodology the Court
described above—it “calculated . . . the average increase (over all relevant hospitals) in hospital
operating costs; and it divided that figure by the increase in the ‘market basket’ . . . for the same
time period. . . . HHS then averaged that number over three prior years.” Pls.’ Summ. J. Mem. at
16 (citing 71 Fed. Reg. at 48,150). HHS explained that it believed this calculation was “more
accurate and stable than the commenters’ methodology because it takes into account the costs per
discharge and the market basket percentage increase when determining a cost adjustment factor.”
71 Fed. Reg. at 48,151. Using this method, the final FY 2007 rule adopted a fixed loss threshold
of $24,475, “$1,055 lower than the $25,530 threshold from the proposed rule.” Id.
The final rule’s fixed loss threshold of $24,475, calculated using a CCR adjustment
factor, was a logical outgrowth of the proposed rule’s suggestion of a fixed loss threshold of
$25,530, calculated using CCRs from the most recent available previous year’s data. A final rule
is a logical outgrowth of a proposed rule “[i]f interested parties ‘should have anticipated’ that the
change was possible, and thus reasonably should have filed their comments on the subject during
the notice-and-comment period . . . .” Am. Coke & Coal Chems. Inst. v. EPA, 452 F.3d 930,
938–39 (D.C. Cir. 2006) (cleaned up). “It generally is not a violation of notice and comment
requirements to amend a proposed rule in response to a comment.” Solite Corp. v. EPA, 952
F.2d 473, 496 (D.C. Cir. 1991); see Select Specialty Hosp.-Akron, LLC v. Sebelius, 820 F. Supp.
2d 13, 23 (D.D.C. 2011) (“There is no requirement that an agency ‘select a final rule from
among the precise proposals under consideration during the comment period.’” (quoting Sierra

15

Club v. Costle, 657 F.2d 298, 352 (D.C. Cir. 1981))). Thus, the agency need not “assiduously lay
out every detail of a proposed rule for comment.” Horsehead Res. Dev. Co. v. Browner, 16 F.3d
1246, 1268 (D.C. Cir. 1994). Indeed, “[t]he whole rationale of notice and comment rests on the
expectation that the final rules will be somewhat different—and improved—from the rules
originally proposed by the agency.” City of Stoughton v. EPA, 858 F.2d 747, 753 (D.C. Cir.
1988). The D.C. Circuit’s notice-and-comment cases “focus . . . primarily on whether the final
rule changes critically from the proposed rule . . . . The question is typically whether the
agency’s final rule so departs from its proposed rule as to constitute more surprise than notice.”
Air Transp. Ass’n of Am. v. F.A.A., 169 F.3d 1, 7 (D.C. Cir. 1999) (emphasis added); see Long
Island Care at Home, Ltd., 551 U.S. at 174 (noting that the object of the logical outgrowth
doctrine “is one of fair notice”).
In the notice of proposed rulemaking for FY 2007, HHS opened for comment its proposal
to use a fixed loss threshold of $24,475 as a means of hitting its 5.1 percent target; it also
proposed for comment its method of achieving that target, which included calculating the yearon-year average annualized rate of change for charges and using the cost-to-charge ratios from
the most recent available historical data. 71 Fed. Reg. at 24,149–50. It explained, in part, that
this method sought to account for the trend of declining cost-to-charge ratios. Id. at 24,150 (“We
note that the case-weighted national average cost-to-charge ratio declined by approximately 1
percent from the March 2005 to the December 2005 update of the Provider-Specific File.
Hospital charges continue to increase at a steady rate of growth between 7 and 8 percent over
each of the last 2 years, resulting in a decline to the cost-to-charge ratios that are used to compute
the outlier threshold. Using lower cost-to-charge ratios from the December 2005 ProviderSpecific File, in combination with the FY 2005 MedPAR claims and inflated charges,

16

contributes to a higher proposed outlier threshold for FY 2007 compared to FY 2006.”).
Interested hospitals should have anticipated that HHS’s proposed threshold and its method of
calculating it might change in some way, including via refinement of the CCR-projection
element of the calculation. See Long Island Care at Home, Ltd., 551 U.S. at 175 (“Since the
proposed rule was simply a proposal, its presence meant that the Department was considering the
matter.”). Any hospital that wished to suggest that the threshold was too high to hit the target, or
to suggest any adjustment to the projection methodology, was on notice of its opportunity to do
so. And that is exactly what hospitals did. HHS’s final methodology built upon these
comments. 71 Fed. Reg. at 48,149–50 (noting in the final rule commenters’ concern[s] that the
proposed threshold was too high and their suggestions for “further refining the outlier
methodology” and agreeing that it would “increase the precision of [HHS’s] model” to “account
for the rate of change in the relationship in costs and charges”—i.e. the precise nature of the
trend of declining CCRs—by “applying an adjustment factor to the CCRs”). “Commenters
clearly understood that” how best to project CCRs to account for the trend of decline was “under
consideration, as the agency received comments [on the method] from several sources.” See
Appalachian Power Co. v. EPA, 135 F.3d 791, 816 (D.C. Cir. 1998).
This case is similar to Solite Corp. v. EPA, 952 F.2d at 485 (per curiam), which
concerned an EPA rulemaking geared at determining whether certain types of industrial wastes
were subject to a particular regulatory category. A waste qualified for the category if it was both
“high volume” and “low hazard.” See id. at 480. In the notice of proposed rulemaking, the EPA
relied on a combination of plant-specific and industry wide data to propose that a solid waste
would qualify as “high volume” if its annual output crossed the threshold of 50,000 metric tons
per year and a liquid waste would qualify if its annual output crossed the threshold of 1.5 million

17

metric tons per year. Id. But during the comment period, “more recent and complete” data
became available that included “detailed information about volumes and specific types of wastes
generated” at relevant facilities. Id. at 481 (citation omitted). In response, the EPA “revised” its
analysis, and the final rule proposed an annual volume threshold for qualification of 45,000
metric tons for solid wastes and one million tons for liquid wastes. Id. (citation omitted). EPA’s
analysis had changed between the proposed and final rule, but its methodology “[had] not
change[d] significantly.” Id. at 485. The new data had allowed EPA to employ “a more precise
quantitative measure based on more complete information” and allowed EPA to adjust in
response to comments critical of the data used in the proposed rule. Id. The D.C. Circuit upheld
the EPA’s rule against a notice-and-comment challenge. Id.
Similarly, in the instant case, HHS proposed a figure (the fixed loss threshold for FY
2007) and explained its methodology for projecting it, including its effort to account for the trend
of decreasing CCRs. Then, in response to analysis received during the comment period, it
adjusted one piece of its overall method. HHS undoubtedly changed its methodology—it did not
“remain constant”—but it did so in a discrete way that “confirmed,” and built upon, its
announced plan to account for declining CCRs. Id. at 485. Where, as here, “an agency’s
analytic task begins rather than ends with a set of forecasts, sound practice would seem to dictate
disclosure of those forecasts so that interested parties can comment upon the conclusions
properly to be drawn from them.” Air Transp. Ass’n of Am., 169 F.3d at 8 (quoting Indep. U.S.
Tanker Owners Comm. v. Lewis, 690 F.2d 908, 926 (D.C. Cir. 1982)).6 HHS did so here, and

6

Air Transport Association of America involved an F.A.A.-specific procedure that is
“similar to the notice and comment procedure for informal rulemaking under the Administrative
Procedure Act,” so the D.C. Circuit applied logical outgrowth principles drawn from APA cases.
169 F.3d at 6–7.

18

adjusted its forecast methodology in a discrete way that was in line with, though not identical to,
suggestions it received from commenters. Compare Am. Coke & Coal Chems. Inst., 452 F.3d at
940 (“Given that the commenters ‘had a fair opportunity to present their views on how the
industry ought to be subcategorized’ and that the choice to merge the subcategories was a
foreseeable result of the EPA’s solicitation of comment on rationality of the subcategory scheme,
there was no failure of notice or opportunity to comment even though parties may not have been
able to predict how the EPA would choose to act upon their comments.” (quoting BASF
Wyandotte Corp. v. Costle, 598 F.2d 637, 642–46 (1st Cir. 1979))), with CSX Transp., Inc. v.
Surface Transp. Bd., 584 F.3d 1076, 1081–82 (D.C. Cir. 2009) (holding that a final rule
permitting ratemaking parties to rely on four years of comparison data for a certain benchmark
was not a logical outgrowth of a proposal to allow for one year of comparison data, because even
though the rule was not “a complete turnaround from the NPRM,” there was “no way
commenters . . . could have anticipated” that this “particular aspect[] of [the] . . . proposal [was]
open for consideration (citation omitted)).
It is worth pausing further over this point: Plaintiffs are correct that it may not have been
possible for hospitals to predict precisely how HHS would respond to its solicitation for
comments on the CCR projection piece of its methodology for setting the FY 2007 outlier
threshold. They may not have foreseen, for example that HHS would take the three-year average
of the increase in hospital operating costs compared to the market basket increase, rather than, as
some commenters suggested, simply deriving the cost inflation factor from “the annual rate of
increase in the cost per discharge” in recent years. See 71 Fed. Reg. at 48,150. But case law
suggests a distinction between major shifts in direction or policy between proposal and rule,
which are not logical outgrowths, and mere refinements or adjustments consistent with the

19

direction of a proposed policy, which are. CSX Transp., Inc, 584 F.3d at 1081 (noting that “D.C.
Circuit cases finding that a rule was not a logical outgrowth have often involved situations where
the proposed rule gave no indication that the agency was considering a different approach, and
the final rule revealed that the agency had completely changed its position,” and citing cases in
which the final rule “adopt[ed] a maximum velocity cap where a minimum was proposed” and
adopted a “completely different reading of a set of regulatory standards” from the reading
proposed (emphasis added)); Air Transp. Ass’n of Am. v. C.A.B., 732 F.2d 219, 224 (D.C. Cir.
1984) (final rule was a logical outgrowth because the “critical elements of the proposal did not
change”). Thus, in contrast to the D.C. Circuit’s conclusion in Solite Corp. that the change in the
high volume threshold did not violate notice-and-comment requirements because it was merely
the result of EPA’s use of better data “to check or confirm prior assessments,” 952 F.2d at 485,
the same panel held in a later portion of the opinion that the EPA’s final-rule conclusion that a
particular waste would be excluded from the regulatory category at issue was not a logical
outgrowth of its proposed rule, which had listed that waste as a candidate for inclusion in the
regulatory category. The proposed rule had not listed the waste on a separate list of proposed
wastes to be excluded, and did not explain the methodology EPA eventually used to exclude the
waste. Id. at 498–500; compare Husqvarna AB v. EPA, 254 F.3d 195, 199, 203 (D.C. Cir. 2001)
(final rule’s imposition of a four-year implementation period to meet new emission standards
was a logical outgrowth of proposed rule’s five-year period, where EPA decided during
comment period that “‘rapid technological advances’ . . . warranted a more expeditious
implementation”), and Select Specialty Hosp.-Akron, LLC, 820 F. Supp. 2d at 23–24 (rule
imposing discharge limits on certain hospitals-within-hospitals in the first year of a transition
period for implementing a new policy limiting hospitals-within-hospitals’ admissions from their

20

host hospitals to 25 percent of their discharges was a logical outgrowth of proposals that did not
mention the first-year requirements, because these were “merely an implementing mechanism for
the 25 percent rule,” which had been “the subject of extensive public comment”), with Allina,
746 F.3d at 1109 (final rule was not a logical outgrowth because it represented “a volte-face”
from the proposal with “enormous financial implications”), and Air Transp. Ass'n of Am., 169
F.3d at 7–8 (final order partially approving airport’s application to impose a passenger facility
charge did not comply with notice-and-comment procedures where the approval was based on a
projection of passenger increase not included in the initial proposal that was “an order of
magnitude greater” than the increase the initial proposal relied upon and was “a measure of
something completely different”).
HHS’s adjustment of its method of determining a single component of its overall project
of estimating the outlier threshold that would result in outlier payments equal to 5.1 percent of
total payments for the final year, see Billings Clinic, 901 F.3d at 307–09, which adjustment
contributed in part to the final rule’s downward revision of the proposed outlier threshold by
$1,055, fits more neatly into the refinement/adjustment category. See Air Transp. Ass’n of Am.,
732 F.2d at 225 n.12 (concluding that a final fee schedule was a logical outgrowth of a proposal
in part because the changes in fees reflected on the final schedule could not “be considered
major”). Owner-Operator Indep. Drivers Ass’n v. Fed. Motor Carrier Safety Admin., 494 F.3d
188, 202 (D.C. Cir. 2007) at first glance seems to undercut this conclusion, but upon closer
inspection confirms it. There, the D.C. Circuit held that the Federal Motor Carrier Safety
Administration (“FMCSA”) violated the notice-and-comment requirement when it promulgated
a final rule governing rest requirements for truck drivers. Id. The final rule employed a model to
estimate truck drivers’ fatigue by using “time-on-task multipliers,” which sought to account for

21

the fact that time spent driving is more fatiguing than time spent resting. Id. at 200–01. The
flaw: though the FMCSA’s model was an “update” to a safety model it had used in a previous
rule and disclosed in the proposal, the “nature of the update” was to use an “entirely new”
methodology, namely the use of time-on-task multipliers, which were “an integral part” of the
model. Id. at 201.
Here, HHS similarly updated the methodology of a disclosed model between the
proposed and final rule, but HHS’s comment-prompted refinement of a discrete (though
important) factor in a multi-part calculation renders “the nature of the update,” id., different in
kind than FMCSA’s wholesale methodological replacement in Owner-Operator Independent
Drivers Association. For one thing, this portion of the Owner-Operator Independent Drivers
Association opinion did not analyze whether the final rule was a logical outgrowth of the
proposed rule; it instead engaged in the related, but distinct, inquiry of whether the methodology
was “critical factual material” that the agency had to disclose for comment. Id. at 199. More
fundamentally, the D.C. Circuit’s key conclusion—that “[a]lthough interested parties may have
known that FMCSA would incorporate time-on-task effects into its crash-risk model, they had no
way of knowing that the agency would calculate the impact of time on task in the way that it
did”—turned on the conclusion that there was “no way” for commenters “to foresee” four
specific elements of the agency’s methodology, three of which contradicted the approach taken
in supporting studies in the administrative record. Id. at 202. It was only “in light of these
undisclosed elements” that the court concluded that the model had not been “made public in the
proceeding and exposed to refutation.” Id. “Moreover,” the court reasoned, “the addition of the
time-on-task element to the model was not a minor modification used to check or confirm prior
analyses: it constituted the agency’s response to an important defect in its previous methodology

22

identified” in a previous D.C. Circuit decision. Id. at 201. As discussed, commenters foresaw
the possibility that HHS would apply an adjustment factor to CCRs in the FY 2007 final rule,
and though they did not precisely predict HHS’s exact final method for deriving the adjustment
factor, the final method was in line with commenter proposals. Though it is a close question, the
FY 2007 final rule’s refinement of a single element of its method of projecting the fixed loss
threshold was more like a “minor modification used to check or confirm prior analyses,” than the
introduction of an “entirely new” methodology held impermissible in Owner-Operator
Independent Drivers Association, id.
The bottom line is that the $1,055 difference between the proposed and final fixed loss
thresholds, a result at least in part of HHS’s change to its method of projecting CCRs, was
“reasonably foreseeable,” Long Island Care at Home, Ltd, 551 U.S. at 175 , so it did not require
the opening of a new round of notice and comment. See Weyerhaeuser Co. v. Costle, 590 F.2d
1011, 1031 (D.C. Cir. 1978) (an agency “need not subject every incremental change in its
conclusions after each round of notice and comment to further public scrutiny before final
action”).
2. HHS was not required to disclose Attachment A during the FY 2007 Rulemaking.
The Court’s conclusion that the FY 2007 final rule was a logical outgrowth of the
proposed rule dictates the further conclusion that Plaintiffs’ Portland Cement argument against
the FY 2007 rule must fail as well. The D.C. Circuit’s Portland Cement or “critical-material”
doctrine, Post Acute Med. at Hammond, LLC v. Azar, 311 F. Supp. 3d 176, 184–85 (D.D.C.
2018), provides that “[u]nder APA notice and comment requirements, among the information
that must be revealed for public evaluation are the technical studies and data upon which the
agency relies in its rulemaking.” Banner Health, 867 F.3d at 1336 (citation omitted); but see Am.
Radio Relay League, Inc., 524 F.3d at 245–47 (Kavanaugh, J. concurring in part, concurring in
23

the judgment in part, and dissenting in part) (concluding that the Portland Cement doctrine is
inconsistent with the Supreme Court’s holding in Vt. Yankee Nuclear Power Corp. v. Natural
Res. Def. Council, Inc., 435 U.S. 519, 524 (1978) that courts may not impose upon agencies
procedural requirements beyond those found in the APA’s text). Plaintiffs invoke this doctrine
to argue that HHS should have provided the opportunity to comment on “Attachment A,” a
document HHS eventually certified for inclusion in the administrative record in this action for
the final FY 2007 rule. AR (FY 2007) at 14403. According to Plaintiffs’ description of the
document, Attachment A was important to HHS’s decision in the final FY 2007 rule to average
hospital costs compared to general inflation over three years (rather than some other period),
because it shows that HHS ran calculations using different averaging periods:
Attachment A reveals that HHS carried out similar calculations using those other
alternative models. In the first table in the document, the column labeled “Oper
MB Increase” is evidently the “market basket” (the “MB”) for a given year. AR
14403. The column labeled “Operating Cost per Discharge Increase” is presumably
the average hospital operating cost increase over the same year. “Cost/MB” is the
number produced by dividing them . . . . “Mean Ratio to 2004” is . . . the average
of the “Cost/MB” figures for the number of years specified in the “Num Years in
Mean” column. The final column is the result of multiplying the “Mean Ratio to
2004” by the “Oper MB Increase” figure for 2005.
Pls.’ Summ. J. Mem. at 17.7 Thus, Plaintiffs appear to acknowledge (and the document
admits of no other conclusion) that Attachment A is relevant only to calculation of the

7

The Court reproduces the described portion here, AR (FY 2007) 14403:

24

adjustment factor to be applied to CCRs (specifically the selection of a three-year average)—a
choice the Court has already concluded was a logical outgrowth of the proposed rule, and
therefore a choice on which HHS was not required to allow further comment. If HHS was not
required to submit its adjustment-factor method for comment, it follows that it was not required
to submit documents underlying that method for comment. See Air Transp. Ass’n of Am., 732
F.2d at 224 (rejecting APA challenge to agency reliance on internal staff studies not publicly
available during the comment period because the “critical elements of the proposal did not
change, and the final rule was a ‘logical outgrowth’ of the proposed rule”); cf. Bldg. Indus. Ass’n
of Superior Cal. v. Norton, 247 F.3d 1241, 1246 (D.C. Cir. 2001) (“[A] final rule that is a logical
outgrowth of the proposal does not require an additional round of notice and comment even if the
final rule relies on data submitted during the comment period.”); City of Stoughton, 858 F.2d at
753(agency was not required to solicit comment on a study that came to its attention only after
the publication of a proposed rule where the study grounded a final-rule conclusion that was a
logical outgrowth of the proposal because it confirmed the proposal’s conclusion); Chamber of
Com. of U.S. v. SEC, 443 F.3d 890, 900, 903–04 (D.C. Cir. 2006) (distinguishing between

25

“entirely new information critical to the agency’s determination” and “supplementary
information,” including data that “clarif[ies], expand[s], or amend[s] other data that has been
offered for comment,” and noting that absent “a showing of prejudice by an interested party,” the
agency need not allow “further opportunity for comment[] provided that the agency’s response
constitutes a ‘logical outgrowth’ of the rule proposed” (citations omitted and cleaned up)). This
result makes sense, as the purpose of the critical material doctrine is to facilitate comment on a
proposed rule. See Banner Health, 867 F.3d at 1336 (the doctrine “allows for useful criticism,
including by enabling commenters to point out where information is erroneous or where the
agency may be drawing improper conclusions” (citations omitted and cleaned up)). By hewing
sufficiently closely to the proposal to render its final rule a logical outgrowth, the agency has
already complied with its obligation to allow a sufficient opportunity for comment.
B. FY 2008–2013 Rules
1. The Court may consider Attachment A and the associated memorandum in order to evaluate
Plaintiffs’ critical-material challenge to the FY 2008–2013 rules.
Plaintiffs next expand their critical material argument against the FY 2007 rule to assert
that HHS violated the Portland Cement doctrine when it failed to disclose Attachment A and the
accompanying memorandum for comment during the rulemakings for FYs 2008–2013. Pls.’
Summ. J. Mem. at 20–22. As Plaintiffs acknowledge, Attachment A and the associated
memorandum (entitled “Change to Outlier Methodology”) are not currently part of the
administrative records for these rules, and their argument on this point “depends on the
assumption that Attachment A and the associated memo are in the record for the FY 2008
through 2013 rules.” Id. at 20 n.12. Therefore, before reaching the merits of this argument, the
Court must first consider, in part, Plaintiffs’ motion to complete the administrative records. In
this motion, Plaintiffs argue that Attachment A and the Change to Outlier Methodology

26

memorandum should be part of the FY 2008–2013 administrative records for all purposes
because HHS either actually considered these documents during these rulemakings or
negligently failed to do so. Pls.’ Mot. Complete Admin. R. at 5–12. In the alternative, they
argue that the Court should at least consider these documents for the limited purpose of
determining whether HHS violated Portland Cement by failing to disclose them during the FY
2008–2013 rulemakings. Id. at 13–16. The Court concludes that it may consider the documents
at least in order to evaluate the procedural Portland Cement claim and accordingly grants the
motion to complete in part.8

The parties have vigorously briefed whether Plaintiffs’ motion to complete the
administrative record by adding Attachment A and the associated memorandum is timely, but, as
the Court sees it, the issue is relatively straightforward. On March 31, 2020, the Court ordered
the Secretary to produce the memorandum to Plaintiffs. Mot. Dismiss Op. at 33 (“The Court will
direct the Secretary to disclose to Plaintiffs the full document with which “Attachment A” is
associated.”). It further held that “[o]nce such document is disclosed, Plaintiffs may, if
appropriate, renew their argument that this document should be part of the administrative
record.” Id. Thus, the Court contemplated a departure from its previous scheduling order, which
set a deadline for supplementation motions of June 5, 2019. ECF No. 137. To be sure, the Court
said this in the context of evaluating Plaintiffs’ assertion that the memorandum was before the
agency during the FY 2007 rulemaking, and therefore arguably did not refer to a motion to
complete the administrative records for FYs 2008–2013. But the Court said Plaintiffs could
renew their “argument,” a term broader than the 2007-specific motion, and, given that the
opinion dealt with challenges to a range of different years’ rules, the use of the word “record” is
fairly read to refer to the entire record before the Court, even though that record is technically
comprised of individual records for various rulemakings. Due to intervening litigation about the
scope of the required disclosure, the Secretary did not produce the memorandum until April 13,
2021. Though Plaintiffs may have been able to surmise it on their own previously, this
production provided—for the first time— record evidence that Attachment A was produced by
the actuary’s office in order to develop the cost inflation adjustment factor method disclosed in
the final FY 2007 rule. Shortly thereafter, in a joint status report, Plaintiffs informed the Court
that they might “need to file a further motion regarding the content of the administrative
records . . . .” ECF No. 179 at 2. About three months after the production, on July 16, 2021,
Plaintiffs moved to add Attachment A and the memorandum to the FY 2008–2013 administrative
records on the theory that the FY 2007 rule relied on these actuary documents, and the
subsequent rules referenced the FY 2007 rule’s actuarial analysis as the basis for their own
methodologies.
8

Moreover, even though this motion to complete came the same day as Plaintiffs’ motion
for summary judgment, the Secretary had two months to account for the filing before submitting
27

The “record rule” generally prohibits the consideration of evidence outside the
administrative record in APA actions: “[I]t is black-letter administrative law that in an APA case,
a reviewing court should have before it neither more nor less information than did the agency
when it made its decision.” Hill Dermaceuticals, Inc. v. FDA, 709 F.3d 44, 47 (D.C. Cir. 2013)
(cleaned up). However, a court may make an exception when “the procedural validity of the
agency’s action remains in serious question.” CTS Corp. v. EPA, 759 F.3d 52, 64 (D.C. Cir.
2014) (citation omitted); see United Student Aid Funds, Inc. v. Devos, 237 F. Supp. 3d 1, 4
(D.D.C. 2017) (describing the exception for cases in which “the district court cannot determine
from the administrative record whether the agency complied with its procedural obligations”
(first citing Esch v. Yeutter, 876 F.2d 976, 991 (D.C. Cir. 1989), and then citing CTS Corp, 759
F.3d at 64)). Still, this exception applies “at most” to challenges to “gross procedural
deficiencies” (citation and emphasis omitted). CTS Corp., 759 F.3d at 65.
If Plaintiffs are right that Attachment A and the Change to Outlier Methodology
memorandum were critical material to the FY 2008–2013 rulemakings and HHS withheld them
from commenter scrutiny, then HHS’s notice-and-comment procedures were grossly deficient.
See Portland Cement, 486 F.2d at 392 (describing the failure to disclose critical material as a

his own summary judgment motion in September. Therefore, unlike in Banner Health v.
Burwell, 126 F. Supp. 3d 28, 60 (D.D.C. 2015), aff’d in relevant part, Banner Health, 867 F.3d
at 1336, in which the court considered it “tardy” to file a motion for consideration of extra-record
material on the same day that both plaintiffs’ and defendant’s motions for summary judgment
were due, the Secretary has not been prejudiced. The D.C. Circuit held that this conclusion was
within the scope of the district court’s discretion in part because the “delay denied HHS the
opportunity to treat the comment as part of the administrative record in preparing its motion for
summary judgment.” Banner Health, 867 F.3d at 1336. That is not the case here. “Given the
general judicial preference for resolving motions on their merits rather than dismissing them on
technicalities,” Niedermeier v. Off. of Baucus, 153 F. Supp. 2d 23, 27 (D.D.C. 2001), under these
circumstances, the Court finds that the motion to complete the 2008–2013 rules is timely and
will consider its merit.

28

“critical defect”); Solite Corp., 952 F.3d at 484 (describing such a failure a “serious procedural
error” (citation omitted)); Am. Radio Relay League, Inc., 524 F.3d at 237 (“It would appear to be
a fairly obvious proposition that studies upon which an agency relies in promulgating a rule must
be made available during the rulemaking in order to afford interested persons meaningful notice
and an opportunity for comment. It is not consonant with the purpose of a rule-making
proceeding to promulgate rules on the basis of inadequate data, or on data that, [to a] critical
degree, is known only to the agency.” (cleaned up)); Ctr. for Biological Diversity v. U.S. Army
Corps of Eng’rs., No. 20-CV-103, 2020 WL 5642287, at *14–15 (D.D.C. Sept. 22, 2020)
(granting plaintiffs permission to rely on extra-record evidence in order to argue that an agency
failed to comply with applicable notice-and-comment procedures). The Secretary objects on the
ground that CTS Corp. “held that an alleged violation of the [Portland Cement] doctrine did not
justify consideration of extra-record evidence.” Def.’s Admin. R. Opp’n at 13. But that is a
misstatement of the case’s holding—the plaintiff attempted to introduce record evidence not to
show a procedural notice-and-comment defect, but rather to challenge the substance of the
agency’s decision. CTS Corp., 759 F.3d at 63–65 (characterizing the plaintiff’s attempt in a
footnote to frame its argument as procedural as “conclusory,” noting that plaintiff did not seek a
remedy appropriate for a procedural violation, and observing that the plaintiff “could have
pursued a procedural challenge arguing that the EPA’s failure to include [certain] data in the
record at the promulgation stage required that it be afforded an additional opportunity to
comment on the data” but that it did not do so (emphasis added)). Moreover, unlike in this case,
the CTS Corp. plaintiff asked the Court to rely on “its own expert’s newly created analysis”
responding to data in the final administrative record, not to consider data that was within the
agency’s files during the rulemaking and allegedly should have been disclosed for comment. Id.

29

at 65. The Court will consider Attachment A and the Change to Outlier Methodology
memorandum in evaluating Plaintiffs’ critical material challenge to the FY 2008–2013 rules.
2. HHS was not required to disclose Attachment A and the associated memorandum during the
FY 2008–2013 rulemakings.
Plaintiffs’ critical material challenge is that together, the memorandum and Attachment A
were critical factual material for the FY 2008–2013 rules because they represented HHS’s
“analysis of alternative methodologies”; that is, they showed HHS’s exploration of the effect of
using averaging periods other than three years on the CCR adjustment factor. Pls.’ Summ. J.
Mem. at 20–22. According to Plaintiffs, Attachment A’s revelation that different averaging
periods resulted in widely varying adjustment factors belies HHS’s assertions in the FY 2008–
2013 rules that its adjustment factor method was “accurate and stable.” Id. at 23–24 (citation
omitted).
Once again, under Portland Cement, the agency must “identify and make available
technical studies and data that it has employed in reaching the decisions to propose particular
rules. An agency commits serious procedural error when it fails to reveal portions of the
technical basis for a proposed rule in time to allow for meaningful commentary.” OwnerOperator Indep. Drivers Ass’n, Inc., 494 F.3d at 199 (cleaned up). This disclosure rule does not
extend to all data in an agency’s files; rather, it applies to “the ‘most critical factual material’
used by the agency.” See Chamber of Com. of U.S., 443 F.3d at 900 (citation omitted). Thus,
when “staff reports” are critical to the agency’s rule, the agency must submit them for comment.
See Am. Radio Relay League, Inc., 524 F.3d at 236 (citation and emphasis omitted). Assuming
without deciding that HHS considered Attachment A and the Change to Outlier Methodology
memorandum during the FY 2008–2013 rulemakings, as opposed to only during the FY 2007
rulemaking—something the Secretary denies, Def.’s Summ. J. Mem. at 21–22—the Court

30

concludes that neither the analysis of alternative averaging periods found in Attachment A nor
the background material found in the Change to Outlier Methodology memorandum were critical
material the agency was required to publish during the FY 2008–2013 rulemakings.
In the rulemakings for FY 2008–13, HHS proposed to use the same fixed loss threshold
projection methodology, including the same CCR adjustment-factor method, that it had used in
the FY 2007 final rule:
As discussed in the FY 2007 final rule (71 FR 48150), we worked with the Actuary
to derive the methodology described below to develop the CCR adjustment factor.
For FY 2008, we are proposing to use the same methodology by using the operating
cost per discharge increase in combination with the final updated market basket
increase determined by Global Insight, Inc., as well as the charge inflation factor
described above to estimate the adjustment to the CCRs. By using the market
basket rate-of-increase and the increase in the average cost per discharge from
hospital cost reports, we are using two different measures of cost inflation. For FY
2008, we determined the adjustment by taking the percentage increase in the
operating costs per discharge from FY 2004 to FY 2005 (1.0529) from the cost
report and dividing it by the final market basket increase from FY 2005 (1.043).
We repeated this calculation for 2 prior years to determine the 3-year average of
the rate of adjusted change in costs between the market basket rate-of increase and
the increase in cost per case from the cost report (FY 2002 to FY 2003 percentage
increase of operating costs per discharge of 1.0721 divided by FY 2003 final market
basket increase of 1.041, FY 2003 to FY 2004 percentage increase of operating
costs per discharge of 1.0624 divided by FY 2004 final market basket increase of
1.04). For FY 2008, we averaged the differentials calculated for FY 2003, FY 2004,
and FY 2005 which resulted in a mean ratio of 1.0203. We multiplied the 3-year
average of 1.0203 by the 2006 market basket percentage increase of 1.0420, which
resulted in an operating cost inflation factor of 6.32 percent or 1.0632. We then
divided the operating cost inflation factor by the 1-year average change in charges
(1.0726) and applied an adjustment factor of 0.9912 to the operating CCRs from
the Provider-Specific File.
See 72 Fed. Reg. 24,680, 24,837 (May 3, 2007), AR at 00159 (FY 2008 Proposed Rule). It then
adopted this methodology in each final rule for FY 2008–13. See 72 Fed. Reg. 47,130, 47,419
(Aug. 22, 2022); Pls.’ Summ. J. Mem. at 20 n.13 (collecting citations for each year’s final rule).
Thus, each proposal referenced the 2007 rulemaking, and, according to Plaintiffs, incorporated
HHS’s analysis of Attachment A and the covering memorandum during the 2007 rulemaking.

31

During the comment period for the FY 2008 rule, one commenter suggested that HHS’s
method of deriving the CCR adjustment factor was “unnecessarily complicated and [did] not
lead to a more accurate” result; the commenter instead suggested “a methodology that uses
recent historical industry wide average rate of change, similar to the methodology used to
develop the charge inflation factor.” 72 Fed. Reg. at 47,418. The commenter does not appear to
have specifically objected to the selection of a three-year averaging period, except insofar as this
choice was one element of the overall adjustment-factor calculation method the commenter
criticized as too complex. See id.; AR at 728. HHS rejected this complaint on the ground that
incorporating two sources of inflation (cost per discharge and market basket increase) would
improve accuracy and stability:
[W]e believe [our] calculation of an adjustment to the CCRs is more accurate and
stable than the commenter’s methodology because it takes into account the costs
per discharge and the market basket percentage increase when determining a cost
adjustment factor. There are times where the market basket and the cost per
discharge will be constant, while other times these values will differ from each
other, depending on the fiscal year. Therefore as mentioned above, using the
market basket in conjunction with the cost per discharge uses two sources that
measure potential cost inflation and ensures a more accurate and stable cost
adjustment factor.
72 Fed. Reg. 47,130, 47,418 (Aug. 22, 2007). The records for the FY 2009–2012 rules include
very similar comments and responses. Pls.’ Summ. J. Mem. at 13 (citing 73 Fed. Reg. 48,434
(Aug. 19, 2008) 48,763, AR 4928 (FY 2009 rule); 74 Fed. Reg. 43,754, 44,008 (Aug. 27, 2009),
AR 7080 (FY 2010 rule); 75 Fed. Reg. 50,042, 50,427 (Aug. 16, 2010), AR 7815 (FY 2011
rule); 76 Fed. Reg. 51,476, 51,792 (Aug. 18, 2011), AR 11757 (FY 2012 rule)). In 2013,
commenters suggested a wider range of alternatives for adjusting CCRs (none of which
specifically took issue with the three-year averaging period). But HHS stuck with the method it
had been using since 2007, explaining that it needed to study the proposed alternative methods
further. 87 Fed. Reg. 53,258, 53,694 (Aug. 31, 2012).
32

As these proposals and exchanges make clear, HHS disclosed in detail during every
rulemaking each step of its methodology for projecting the fixed-loss threshold, and in turn each
step of its sub-methodology for deriving the CCR adjustment factor, including the selection of a
3-year averaging period. Commenters understood the methodology and took advantage of the
opportunity to suggest alternatives. HHS’s full disclosure of the methodology (and its results)
was sufficient to its critical-material disclosure obligations. Cf. Air Transp. Ass’n of Am., 732
F.2d at 224 (holding that a final fee schedule that turned in part on agency staff studies not
available during the comment period complied with notice-and-requirements because the
proposal “both outlined the method by which [the agency] proposed to calculate the fees and
listed the types of fees it proposed to charge”); Indep. U. S. Tanker Owners Comm., 690 F.2d at
926 (“[W]here an agency’s analytic task begins rather than ends with a set of forecasts, sound
practice would seem to dictate disclosure of those forecasts so that interested parties can
comment upon the conclusions properly to be drawn from them.”); Chamber of Com. of U.S, 443
F.3d at 900 (observing that a second round of notice and comment is not required when an
agency “merely supplements information in the rulemaking record” by “internally generating
information using a methodology disclosed in the rulemaking record”).
It is not as if the document Plaintiffs contend should have been disclosed, Attachment A,
was central to the methodology. It relates at most to a discrete piece of the methodology—
averaging the increase in operating costs compared to the market basket increase over three years
as opposed to some other time period—and even then arguably consists mostly of evaluation of
the results of alternative averaging periods HHS did not ultimately propose or select, rather than
analysis of the three-year choice itself. AR at 14403; see Pls.’ Summ. J. Mem. at 17 (identifying
“[w]hat HHS did not disclose” as the fact that “it had done comparable calculations using

33

averages over two years, four years, five years, etc.”). HHS was not required to disclose analysis
of methodologies it did not ever propose or adopt as a rule.
As for the Change to Outlier Methodology memorandum itself, in the redacted form in
which it appears in the record before the Court, it contains only an anodyne overview of the
adjustment factor methodology and the reasoning behind it. To a significant extent, this
discussion duplicates the description HHS made public in the proposed rules:
We requested OACT [Office of the Actuary] assistance in developing an
adjustment to the cost-to-charge ratios. OACT developed a cost inflation factor
that could be used in our model. It combines the rate of increase in cost per case
with the rate of increase in the IPPS market basket.
If charges are increasing faster than costs, the cost-to-charge ratios will decline,
estimates of cost per case and the outlier threshold will be lower than it otherwise
would be in the absence of making such an adjustment. The opposite will occur if
costs are increasing faster than charges. For FY 2007, the charge inflation factor is
slightly higher than the cost inflation factor resulting in a small reduction to the
cost-to-charge ratios.
AR (FY 2007) at 14407. These generalities are not the type of analysis the critical-material
doctrine contemplates. In any event, HHS had publicly explained more or less all of the
reasoning reflected in these paragraphs by the time the FY 2008 proposed rule opened for
comment. See 72 Fed. Reg. 24,680, 24,837 (May 3, 2007), AR (FY 2008) at 00159 (explaining
in the FY 2008 proposed rule that “by using the market basket rate-of-increase and the increase
in the average cost per discharge from hospital cost reports, we are using two different measures
of cost inflation”); 71 Fed. Reg. at 48,150 (explaining in the FY 2007 final rule that HHS
decided to use an adjustment factor in order to “account for the rate of change in the relationship
between costs and charges”).
That HHS publicly disclosed a detailed step-by-step methodology, complete with the
figures yielded at each step, distinguishes this case from the key authorities on which Plaintiffs
rely. In Owner-Operator Independent Drivers Association, Inc., for example, the D.C. Circuit
34

vacated parts of a rule regulating truck-driver rest practices because the agency “fail[ed]” to
“disclose the methodology” of a model that was “central to the agency’s justification for the
rule.” 494 F.3d at 199–200. In particular, the agency failed to disclose its use of certain
multipliers that measured driver fatigue based on time spent driving; these were “an integral
part” of the model. Id. at 201. Here, in contrast, HHS disclosed each step of its methodology,
including the three-year averaging period Plaintiffs contest. Indeed, Plaintiffs do not claim that
HHS failed to disclose its methodology, but rather “its analysis of alternative methodologies.”
Pls.’ Summ. J. Mem. at 20 (cleaned up). Yet they do not cite any authority for the proposition
that, in addition to disclosing “portions of the technical basis for a proposed rule,” OwnerOperator Indep. Drivers Ass’n, Inc., 494 F.3d at 199 (cleaned up and emphasis added), an
agency must also disclose material related to its evaluation of a rule or methodology it ultimately
chose not to pursue and not to propose to the public. Cf. Chamber of Com. of U.S., 443 F.3d at
902 (finding a critical-material violation where undisclosed materials “supplie[d] the basic
assumptions” the agency used to support its cost-benefit analysis); Penobscot Indian Nation v.
U.S. Dep’t of Hous. & Urb. Dev., 539 F. Supp. 2d 40, 49 (D.D.C. 2008) (finding a criticalmaterial violation where the agency failed to disclose an internal analysis that was the
“centerpiece” for the final rule’s rationale where the proposed rule neither referenced the
analysis nor included “at least a summary of the specific data and methodology on which the
analysis relied”).
Similarly, in Shands Jacksonville Med. Ctr. v. Burwell, 139 F. Supp. 3d 240, 247, 265
(D.D.C. 2015), the court held that HHS failed to meet its procedural obligations when it relied on
an undisclosed actuary analysis to estimate that a change in how patient stays were classified
would increase the number of patients classified as inpatients, and, in turn, to reduce hospital

35

compensation for inpatient services. The court framed “[t]he question” as “whether the public
was aware of the methodology the HHS actuaries used to predict the effects of” the classification
change. Id. at 264. The court rejected the Secretary’s contention that the proposed rule
discussed the methodology, because all the proposed rule said was that “the Secretary analyzed
‘FY 2009 through FY 2011 Medicare claims data for extended hospital outpatient encounters
and shorter stay hospital inpatient encounters.’” Id. at 263 (quoting the proposed rule). This did
not reveal “critical assumptions” reflected in an actuarial memorandum used in reaching the
estimate, including that the actuarial analysis accounted only for claims related to certain types
of procedures. Id. at 262–64. Thus, there was “no reason to believe that commenters had any
idea what the actuaries did” with the underlying data. Id. at 264. Here, HHS told the public
exactly what its actuaries did, right down to the selection of a three-year averaging period. The
selection of a three-year period may have been an assumption, but unlike in Shands, HHS
disclosed the choice and opened it to commenter criticism.
American Radio Relay League, Inc. v. FCC is probably the strongest case for Plaintiffs,
but it, too, is meaningfully distinct. In that case, the D.C. Circuit took issue with the fact that the
agency’s “determination [was] based upon a complex mix of controversial and uncommented
upon data and calculations.” 524 F.3d at 237 (citation omitted and cleaned up). The agency had
disclosed parts of studies that it admitted were “a central source of data for its critical
determinations,” but had redacted other parts of the studies. Id. at 238. These hidden portions
“appear[ed] to contain information in tension with the Commission’s conclusion,” id. (citations
omitted and cleaned up); similarly, Plaintiffs contend that Attachment A’s demonstration that the
adjustment factor “could fluctuate significantly” depending on the averaging period chosen “was

36

contrary to HHS’s assertion that the method was accurate and stable.” Pls.’ Summ. J. Mem. at
22.9
Though the American Radio Relay League panel made some sweeping statements during
the course of its discussion, e.g., 524 F.3d at 238 (“It would appear to be a fairly obvious
proposition that studies upon which an agency relies in promulgating a rule must be made
available during the rulemaking in order to afford interested persons meaningful notice and an
opportunity for comment.”), it went on to emphasize the “narrowness of [its] holding,” which
turned in significant part on the fact that “the Commission ha[d] chosen to rely on” the studies
“and to place them in the rulemaking record” while redacting portions that undermined its
position. Id. at 239. In other words, the FCC could not play “hide and seek” by “cherrypick[ing]” only the favorable parts of “a study on which it ha[d] chosen to rely in part.” Id. at
237, 239. This feature was a “critical distinction” in American Radio Relay League, id. at 239,
and nothing like it has occurred here. HHS did not present to the public part of a study in
support of its fixed-loss threshold projection method or its three-year average choice while
carving out Attachment A for exclusion from the study. Indeed, HHS did not support its “more
accurate and stable” assertions by referencing its three-year averaging period at all; read in
context, these turned on the fact that unlike the commenter’s proposed method, HHS’s
adjustment factor took into account “two sources that measure potential cost inflation,” costs per
discharge and the market basket. See, e.g., 72 Fed. Reg. at 47,419 (“[W]e believe [our]

As the Court will explain, it is not convinced that Attachment A undercuts HHS’s
“more accurate and stable” assertions, which in context did not reference the three-year
averaging period but rather compared HHS’s method of accounting for both costs per discharge
and the market basket increase to a commenter proposal that would not have taken into account
“two sources that measure potential cost inflation.” 72 Fed. Reg. at 47,418.
9

37

calculation of an adjustment to the CCRs is more accurate and stable than the commenter’s
methodology because it takes into account the costs per discharge and the market basket
percentage increase when determining a cost adjustment factor. There are times where the
market basket and the cost per discharge will be constant, while other times these values will
differ from each other, depending on the fiscal year. Therefore as mentioned above, using the
market basket in conjunction with the cost per discharge uses two sources that measure potential
cost inflation and ensures a more accurate and stable cost adjustment factor.”). Attachment A,
therefore, is not especially relevant to the primary claim Plaintiffs suggest it undermines. See
Pls.’ Summ. J. Mem. at 22.
In sum, in the 2008–2013 rulemakings, HHS disclosed its overall methodology and each
step thereof. The APA did not require it to disclose an analysis of ultimately unchosen
alternatives to a particular step.
3. The FY 2008 final rule’s use of a positive CCR adjustment factor did not necessitate a further
round of notice and comment.
Plaintiffs attempt one final challenge to the FY 2008 rule in particular. That year, HHS
proposed to use the adjustment-factor method it had developed in the FY 2007 final rule, which,
when applied to available data from the December 2006 update to the provider-specific file,
produced an adjustment factor of 0.9912. 72 Fed. Reg. at 24,837, AR 159. In other words, the
model forecast that CCRs would be “almost 1% lower in FY 2008” than those reflected in the
latest provider-specific file data. Pls.’ Summ. J. Mem. at 18. In the final FY 2008 rule, HHS
used the very same model, but, as was its general practice, fed into the model more recent data
that had become available during the comment period. See 72 Fed. Reg. at 47,418 (describing
use of the March 2007 update to the provider-specific file and noting different figures for various
inputs into the adjustment factor calculation methodology); see generally Pls.’ Summ. J. Mem. at

38

5 n.6. The result was an adjustment factor of 1.0027—a projection, contrary to the projection in
the proposed rule, that CCRs would increase. 72 Fed. Reg. at 47,418. Plaintiffs claim that the
final rule’s application of a positive adjustment factor was an unfair surprise that required further
opportunity for comment, especially given that HHS had in the proposed rule explained that it
sought to reflect the trend of declining CCRs. See Pls.’ Summ. J. Mem. at 18–20 (citing 72 Fed.
Reg. at 24,837, AR 159).
This challenge fails for the same reason Plaintiffs’ critical-material challenges failed:
HHS disclosed each step of the methodology it would use to set the final CCR adjustment factor.
Regulated entities had ample opportunity to comment on and suggest changes to this
methodology That the methodology’s application to the latest data produced a surprising
outcome did not obligate HHS to open a new round of notice-and-comment proceedings. The
very authorities on which Plaintiffs rely for their challenge demonstrate that HHS met APA
procedural requirements here. Plaintiffs stress that “HHS was ‘required to disclose what the
actuaries did with [the] data”—here HHS did so, and the actuaries’ methods did not change
between the proposal and final rule. Pls.’ Summ J. Mem. at 19 (quoting Shands Jacksonville,
139 F. Supp. 3d at 264). Plaintiffs also rely heavily on a case in which an agency expanded,
without warning, from a methodology that used one year of comparison data to one that used
four years of comparison data. CSX Transp., Inc., 584 F.3d at 1081–82. In this case, HHS did
not change its data parameters (e.g., it used a three-year averaging period the entire time); all that
changed was the data the actuaries operated upon, and, accordingly, the model’s numeric output.
Having disclosed its model, HHS was not required to provide specific notice of the model’s
result when applied to updated data. See Am. Coke & Coal Chems. Inst., 452 F.3d at 939 (“It is
perfectly predictable that new data will come in during the comment period, either submitted by

39

the public with comments or collected by the agency in a continuing effort to give the regulations
a more accurate foundation . . . . If data used and disclosed for the interim regulations presented
the issues for comment, then there is no need to seek new comment even though significant
quantitative differences result.” (quoting BASF Wyandotte Corp. v. Costle, 598 F.2d 637, 644–45
(1st Cir.1979)); cf. Post Acute Med. at Hammond, LLC v. Azar, 311 F. Supp. 3d 176, 183
(D.D.C. 2018) (“The Administrative Procedure Act . . . does not require an agency to advise
regulated entities as to the individualized implications of a proposed rule—particularly here,
where the rule merely continued a longstanding policy with updates reflecting new data.”).
Admittedly, HHS did not explain with great clarity in the FY 2008 final rule which
particular data changed and why. Compare, e.g., 72 Fed. Reg. at 47, 418 (noting that the 2006
market basket percentage increase in the final rule was 1.0430), with, e.g., id. at 24, 836 (noting
that the 2006 market basket percentage increase in the proposed rule was 1.0420). But any
failure in explanation would sound as a substantive defect, not a procedural failure to comply
with notice-and-comment requirements. Plaintiffs have not argued that any failure to explain the
data changes reflected in the FY 2008 final rule was arbitrary and capricious. See Pls.’ Summ. J.
Mem. at 26 (“The Hospitals are not here contending . . . that it was arbitrary for the model to
produce a particular CCR adjustment factor (like the increase forecasted for FY 2008.”)).
Accordingly, the FY 2008 rulemaking was procedurally sound.
V. ANALYSIS: SUBSTANTIVE CHALLENGES
In addition to their notice-and-comment challenges, Plaintiffs claim that HHS acted
arbitrarily and capriciously in various ways when it set the fixed loss thresholds for FYs 2007–
2016. See 5 U.S.C. § 706(2)(A). “A rule is arbitrary and capricious if the agency: (1) ‘has relied
on factors which Congress has not intended it to consider,’ (2) ‘entirely failed to consider an

40

important aspect of the problem,’ (3) ‘offered an explanation for its decision that runs counter to
the evidence before the agency,’ or (4) ‘is so implausible that it could not be ascribed to a
difference in view or the product of agency expertise.’” United States Sugar Corp. v. EPA, 830
F.3d 579, 606 (D.C. Cir. 2016) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm
Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). “Nor will [the Court] uphold agency action if it
fails to consider significant and viable and obvious alternatives.” Dist. Hosp. Partners, L.P. v.
Burwell, 786 F.3d 46, 59 (D.C. Cir. 2015). “[T]he party challenging an agency’s action as
arbitrary and capricious bears the burden of proof.” City of Olmsted Falls v. F.A.A., 292 F.3d
261, 271 (D.C. Cir. 2002) (quoting Lomak Petroleum, Inc. v. FERC, 206 F.3d 1193, 1198 (D.C.
Cir. 2000)) (alteration in original). Though Plaintiffs fail to meet this burden for many of their
challenges, they have demonstrated that the FY 2012 and FY 2013 rules were arbitrary and
capricious in certain respects.
1. HHS’s use of a three-year averaging period when calculating the CCR adjustment factor for
FYs 2007–2013 was not arbitrary and capricious.
Plaintiffs first argue that HHS’s decision to average the increase in hospital operating
costs compared to the market basket increase over three years, as opposed to over some other
term of years, was arbitrary and capricious. Plaintiffs say that the possibility of using an
alternative averaging period was a “significant and viable and obvious alternative[],” Pls.’
Summ. J. Mem. at 23 (quoting Dist. Hosp. Partners, L.P, 786 F.3d at 59), and that “the
possibility that a different averaging period might produce a significantly different reduction in
CCRs, compared to the method that HHS did use, [was] surely an important aspect of the
decision-making” that HHS was required to consider, id. The Court disagrees and holds that the
decision to use a three-year averaging period was not significant enough in the context of each
year’s rulemaking to require HHS to consider and explain this particular choice.

41

Plaintiffs have not persuasively argued that whether to average the comparison of
operating cost increases to market basket increases over three years rather than, say, over two or
four years, was a significant aspect of the problem HHS confronted during each rulemaking.
The problem HHS confronted was where to set the fixed loss threshold in order to achieve outlier
payments equal to 5.1 percent of total payments for the coming year. As the Court has
explained, taking the three-year average was but one sub-step in the multi-step process of
projecting the fixed-loss threshold for a given year. Significant problems, by contrast, are those
that are central to the overall objective of the rule. See Util. Solid Waste Activities Grp. v. EPA,
901 F.3d 414, 429–30 (D.C. Cir. 2018) (holding that it was arbitrary and capricious for the EPA
not to consider, when crafting a rule regulating the disposal of toxic wastes produced by
electrical plants, neither “the risks to public health and to the environment before leakage is
detected, nor the harms from continued leakage during the years before leakage is ultimately
halted by retrofit or closure”). Here, whether to account both for operating cost increases and
market basket increases may have been a significant aspect of the problem, but the granular subissue of whether to average that comparison over two, three, or four years was not. See
Michigan v. EPA, 213 F.3d 663, 690–91 (D.C. Cir. 2000) (noting, in the context of rejecting a
challenge to various EPA cost assumptions where the challenger had not explained why the
assumptions were arbitrary and capricious, that courts “generally defer[] to the agency’s
expertise” on “technical details”). This is especially so where plaintiffs have not demonstrated
(or attempted to demonstrate), at least without referencing Attachment A, that an alternative
averaging period would obviously have better helped HHS reach its 5.1 percent target. Pls.’
Summ. J. Mem. at 23 (noting that various alternative periods would have resulted in larger
downward adjustments in CCRs for the FY 2007 and FY 2008 rules without explaining why

42

these adjustments would have been more accurate than the one produced by the three-year period
and conceding that “the point here is not that the Court should determine a two-year or any
particular averaging period would have been superior”).
The Court’s conclusion would hold even were it to consider Attachment A, which shows
that, at least in 2007, HHS was aware that the selection of a particular averaging period could
change the adjustment factor. Plaintiffs rely on Attachment A to demonstrate that “a four-year or
a two-year [averaging period] . . . would, if anything, have produced adjustment factors closer to
the historical trends” and that “different averaging periods could have resulted . . . in lower
thresholds.” Pls.’ Summ. J. Mem. at 22. But matching historical trends is not HHS’s end goal in
setting the fixed-loss threshold; rather, attempting to match historical trends in CCRs is a means
of meeting HHS’s statutory mandate to project payments for the coming year. See, e.g., 71 Fed.
Reg. at 48,151 (explaining the decision to apply an adjustment factor “so that the CCRs we are
using in our simulation more closely reflect the CCRs that will be used in” the coming year); cf.
Banner Health, 867 F.3d at 1355 (“There may well be many non-arbitrary reasons for predicting
that costs and charges in a particular industry will not continue on their current trajectories.”). In
any event, “imperfection alone does not amount to arbitrary decision-making . . . [an] agency
may use a model ‘even when faced with data indicating that it is not a perfect fit.’” Dist. Hosp.
Partners, 786 F.3d at 61–62 (citation omitted). In short, “the Hospitals fail to show why [HHS]
should not receive the deference typically accorded in this context,” Banner Health, 867 F.3d at
1356, and the Court defers to the agency on the technical determination of whether three years
was a an appropriate period over which to compare operating cost and market basket increases.10

10

Plaintiffs also assert that it was arbitrary and capricious for HHS to repeatedly assert
that its methodology was “accurate and stable” “when the outcome of the method could be dialed
up or down by nearly a factor of ten by changing an arbitrary parameter (the averaging period).
43

Now, we do not know why HHS selected a three-year averaging period; there is no
disputing that HHS never explained this choice. But HHS “was not required to . . . ‘consider all
policy alternatives in reaching its decision.’” Dep’t of Homeland Sec. v. Regents of the Univ. of
Cal., 140 S. Ct. 1891, 1914–15 (2020) (quoting State Farm, 463 U.S. at 51). It follows that HHS
was not required to explain choices that did not rise to the level of significant aspects of the
problem or to explain why it did not select alternatives (e.g., averaging over periods other than
three years) that were not “significant and viable and obvious,” Dist. Hosp. Partners, L.P., 786
F.3d at 59. Cf. Texas Mun. Power Agency v. EPA, 89 F.3d 858, 869–70 (D.C. Cir. 1996) (“And
though the EPA did not explain its precise method for calculating a rate based on a statewide
average that was used in this case until after the close of general proceedings before the agency,
the failure of an agency to identify every detail of a process before it is used does not
automatically require judicial interference in matters that must be thought to lie within the
agency’s expertise.”).11

Pls.’ Opp’n Sec’y’s Cross-Mot. and Reply Supp. Mot. Summ. J. at 28 (“Pls.’ Combination
Summ. J. Opp’n & Reply”), ECF No. 193. But as the Court has explained, this takes HHS’s
“accurate and stable” assertions out of context. See supra at 37. These did not relate to the
choice of averaging period or even assert as a general matter that the model was maximally
accurate and stable. Instead, they responded to a particular commenter’s suggestion to rely on
historical trends, and noted that using two inflation measures improved accuracy and stability as
compared to this suggested method. 72 Fed. Reg. at 47,418. Plaintiffs do not explain how the
choice of a three-year averaging period undercuts this specific assertion.
11

Because this claim is the sole remaining one for which consideration of Attachment A
and the Change to Outlier methodology memorandum might make any difference, and because
the Court would reject this claim whether or not Attachment A and the memorandum are in the
administrative records, the Court denies the remaining portion of Plaintiffs’ Motion to Complete
Administrative Records—which seeks to have the documents added to the records for all
purposes, not just for evaluation of Plaintiffs’ procedural claims—as moot. See Pls.’ Mot.
Complete Admin. R. at 5–12; cf. County of Los Angeles, 192 F.3d at 1021–22 (finding it
unnecessary to decide whether an affidavit that had “surfaced for the first time during litigation”
was a “post-hoc rationalization” that should have been stricken from the record because even
were it to consider the affidavit, the Court would still have found against the party offering it).

44

2. HHS’s reasoning for rejecting the use of an “estimate adjustment factor” was sufficient to
support the FY 2011 and FY 2014 rules, but was not sufficient to support the FY 2012 and
FY 2013 rules.
Plaintiffs find some, though not complete, success with their next set of arguments: that
HHS acted arbitrarily in the FY 2011–2014 rulemakings when it refused to adopt a commenter’s
repeated suggestion to apply an “estimate adjustment factor” that the commenter said would
improve the accuracy of the fixed-loss threshold projection. Pls.’ Summ. J. Mem. at 32–34. The
commenter first suggested the application of an estimate adjustment factor—not to be confused
with the adjustment factor applied to CCRs, which the Court has discussed at length—during the
comment period for the FY 2011 rule. The commenter presented data which it said
demonstrated that HHS had fallen short of its 5.1 percent target for outlier payments each year
since 2004. AR (FY 2011) at 09476–77. The commenter suggested that HHS should recognize
this consistent shortfall and quantify it by averaging the percentage shortfall from the 5.1 percent
target over two or three past years. Id. For example, the commenter calculated that the average
shortfall from the 5.1 percent target was .385 percent over 2008 and 2009. Thus, for 2011, the
commenter recommended modelling as usual, but setting a fixed loss threshold that HHS
estimated would result in outlier payments equal to 5.485 percent of total payments (the 5.1
percent target plus the .385 percent estimate adjustment factor). In other words, HHS’s ultimate
goal would remain to achieve 5.1 percent, and it would use the same model it had used in the
past, but it would attempt to adjust for whatever imperfections in the model had caused it to
undershoot by a .385 percent average over the previous period by directing the model to achieve
an artificially high percentage. See id. The commenter further suggested that if HHS ever faced
the problem of its model consistently overshooting its target, it could apply a negative estimate
adjustment factor using the same method. Id. As the Secretary puts it, the commenter suggested

45

that if HHS declined to change specific asserted defects in the model, it should use a “fudge
factor” designed to correct the model’s end result. See Def.’s Summ. J. Mem. at 28.
HHS declined to adopt this proposal for the FY 2011 final rule, suggesting that it may
have been inconsistent with HHS’s statutory mandate:
[U]nder the statute, outlier payments are intended to approximate the marginal cost
of providing care above the outlier fixed-loss cost threshold. Any “estimate
adjustment factor” to the outlier threshold or standardized amount in a given year
to account for “overpayments” or “underpayments” of outliers in other years would
result in us making outlier payments that were not directly related to the cost of
furnishing care in extraordinarily costly cases.
75 Fed. Reg. 50,042, 50,429 (Aug. 16, 2010). Moreover, HHS, perhaps understanding the
commenter’s suggestion to relate to revising past outlier payments in addition to projecting
future ones, reiterated its longstanding policy of not making “retroactive adjustments to outlier
payments to ensure that total outlier payments in a past year are equal to 5.1 percent of total
DRG payments.” Id. HHS further asserted that the commenter’s suggestion did “not lend
greater accuracy to [its] estimate of payments that are 5.1 percent of total DRG payments”
because HHS’s model already “factor[ed] in all payments and policies that would affect actual
payments for the fiscal year at hand . . . .” Id.
The commenter tried again during the FY 2012 rulemaking by repeating its estimateadjustment-factor suggestion in similar terms. It also attempted to assuage HHS’s statutory
concerns by “stat[ing] that [the proposed estimate adjustment factor method] would fulfill the
statutory requirement in section 1886 (d)(5)(A) of the Act that requires that CMS establish
thresholds such that outlier payments will be projected to achieve at least 5.1 percent of DRG
payments and would more closely achieve a result that is fully consistent with the statute.” 76

46

Fed. Reg. 51,476, 51,793 (Aug. 18, 2011) (describing the comment).12 HHS “thank[ed] the
commenter for further explaining their position on this adjustment” but declined to adopt the
adjustment in two short sentences:
Further analysis by CMS is necessary to determine if the commenter’s approach to
applying an “estimate adjustment factor” is appropriate. We will consider the
commenter’s suggestion of applying an “estimate adjustment factor” in future
rulemaking if, based on our analysis, we determine that application of an “estimate
adjustment factor” is appropriate and consistent with the statute.
Id. at 51,794.
Undeterred, the commenter presented the same suggestion again during the FY 2013
rulemaking, complete with data it said showed consistent underpayment. AR (FY 2013) at
13100–01, 13157. But HHS was not swayed, and essentially recycled its 2012 explanation to
again reject the estimate-adjustment-factor proposal in the FY 2013 final rule:
With regard to the comment that CMS implement an “estimate adjustment factor”,
as we stated last year, further analysis by CMS is necessary to determine if the
commenter’s approach to applying such a factor is appropriate. We will consider
the commenter's suggestion to apply an “estimate adjustment factor” (in
conjunction with analyzing the alternative methodologies to adjust the CCRs
discussed above), for future rulemaking if, based on our analysis, we determine that
application of an “estimate adjustment factor” is appropriate and consistent with
the statute.
77 Fed. Reg. 53,258, 53,694 (Aug. 31, 2012).
Finally, the commenter tried again in 2014. AR (FY 2014) at 15864–65. The FY 2014
final rule again did not employ an estimate adjustment factor; this time, HHS did not even

12

Both parties say that the FY 2012 comment can be found at pages 11339, 11404–05 of
the FY 2012 administrative record, but the Court has been unable to locate these pages in either
the original or corrected submissions of the FY 2012 administrative record. However, HHS
described the comment in some detail in the FY 2012 final rule, and this description substantially
matches the content of the comment submitted for FY 2013. Moreover, the parties do not
suggest in their briefing that there are any material differences between the FY 2012 and FY
2013 comments.

47

address the comment. However, HHS made other significant changes to its fixed-loss threshold
projection methodology, which the Court will discuss below. Plaintiffs claim that each of HHS’s
explanations (and the 2014 non-explanation) for rejecting the use of an estimate of adjustment
factor was arbitrary and capricious. Pls.’ Summ. J. Mem. at 32–34. The Court agrees that the
2012 and 2013 explanations, but not the 2011 and 2014 explanations, were insufficient.
FY 2011. To begin with, it is not entirely clear that Plaintiffs are challenging HHS’s FY
2011 explanation for declining to adopt the estimate adjustment factor. They note in their
opening brief that HHS indicated in 2011 that it “thought the method was contrary to the
statute,” and say no more about this before shifting to discussion of the FY 2012 rulemaking in
the very next sentences: “But in the FY 2012 rulemaking, the commenter explained that HHS
had misunderstood the method and it was actually consistent with the statute. HHS accepted that
explanation, thus removing its prior ground for rejecting the proposed method. . . . Why, then,
would HHS not adopt it?” Pls.’ Summ. J. Mem. at 33. Plaintiffs then focus on the assertion that
“[b]y the FY 2012 rulemaking” HHS’s model had missed its target for five years, and complain
that in this context HHS’s vague indications that it needed to address the issue further (and its
failure to say anything at all in the FY 2014 rule) were inadequate—all complaints specific to
HHS’s responses in 2012–2014, but not to relevant to its 2011 response. Id. at 33–34 (emphasis
added). In his combination response brief and memorandum in support of his cross-motion for
summary judgment, the Secretary does not expressly defend HHS’s FY 2011 statutory
explanation—though he does describe it—and instead focuses on arguing that the comment “did
not demonstrate, and did not even purport to demonstrate, that the proposed ‘estimate adjustment
factor’ would operate as a reliable predictor of future differences between projected and actual
payments.” Def.’s Summ. J. Mem. at 29. But then, in their combination response to the

48

Secretary’s cross-motion for summary judgment and reply in support of their own motion for
summary Judgment, Plaintiffs say for the first time that part of HHS’s FY 2011 statutory
reasoning—that the estimate adjustment factor method was inconsistent with the statute because
it “would mean basing the coming year’s forecast on something other than ‘all payments and
policies that would affect actual payments’ for the coming year”—“was arbitrary and
capricious.” Pls.’ Combination Summ. J. Opp’n & Reply at 34 (quoting 75 Fed. Reg. at 50,429).
Use of historical data, Plaintiffs point out in that brief, is core to HHS’s mandate to “‘select
outlier thresholds which, when tested against historical data, will likely produce’” outlier
payments in the targeted range. Id. (quoting County of Los Angeles v. Shalala, 192 F.3d at 1013)
(emphasis added in Plaintiffs’ brief).
Assuming Plaintiffs have properly presented an argument against the FY 2011 response,
the Court rejects it. For one thing, Plaintiffs have not extensively engaged with HHS’s 2011
statutory position; for example, they have not addressed whether the position represented a
reasonable interpretation of an ambiguous statute and is subject to Chevron deference. The
Court will not raise these issues in the first instance. More importantly, Plaintiffs have not
presented any argument that HHS’s other, non-statutory reason for rejecting the estimate
adjustment factor for FY 2011—its assertion that the method would not improve the projection’s
accuracy because HHS’s model already took into account all relevant payments and policies—
was insufficient. 75 Fed. Reg. at 50,429. Plaintiffs’ filings have not convinced the Court that
HHS acted in an arbitrary and capricious manner when it declined to employ an estimate
adjustment factor for FY 2011.13

At this point, the Court has rejected on the merits all of Plaintiffs’ challenges to the
rules for FYs 2007–2011. Therefore, the Court need not and does not address the Secretary’s
alternative argument that even if the challenges to these rules prevail on the merits, certain
13

49

FY 2012 and FY 2013. The result is different for FYs 2012 and 2013. “An agency is
required to consider responsible alternatives to its chosen policy and to give a reasoned
explanation for its rejection of such alternatives.” Spirit Airlines, Inc. v. United States Dep’t of
Transp., 997 F.3d 1247, 1255 (D.C. Cir. 2021) (quoting Am. Radio Relay League, Inc., 524 F.3d
at 242). In FY 2012, HHS was no longer laboring under its apparent misimpression that the
estimate adjustment factor proposal involved revising past outlier payments. See 76 Fed. Reg. at
51,793–94. HHS did say that it still needed to determine whether the estimate adjustment factor
method was “consistent with the statute,” but it did not say whether this concern rested on either
of the statutory points HHS had raised in 2011—which it did not repeat—or on some other
interpretative concern. Id. at 51,794. More generally, the only information contained in HHS’s
response was that it would consider the suggestion “in future rulemaking if, based on our
analysis, we determine that application of an ‘estimate adjustment factor’ is appropriate and
consistent with the statute.” Id. This was insufficient. See Spirit Airlines Inc., 997 F.3d at 1255
(holding that an agency’s statement that it “plan[ned] to assess” an issue commenters had raised
fell “well short of what [was] needed to demonstrate the agency grappled with an important
aspect of the problem before it or considered another reasonable path forward”). Given the
nature of the comments, HHS was required to say something about why it needed more time to
study the statute than the comment period had allowed. The estimate adjustment factor method
went to the core of the issue before HHS (the accuracy of its projections), was relatively simple
(it allowed HHS to retain the same model it had been using), and was accompanied by
concerning data purporting to show that HHS had consistently missed its targets. See id. at

plaintiffs’ “claims for fiscal years 2007 to 2011 . . . are barred by issue preclusion based on
earlier litigation in Banner Health or Billings Clinic.” Def.’s Summ. J. Mem. at 40, 43–45.

50

51,793 (“Based on actual payments determined by the commenter using data analysis, the
commenter asserted that the underpayment has exceeded 0.5 percent in all years except one.”);
see also Thompson v. Clark, 741 F.2d 401, 409 (D.C. Cir. 1984) (an agency’s failure to respond
to a comment is problematic to the extent that “it demonstrates that the agency’s decision was
not ‘based on a consideration of the relevant factors’ (citation omitted)). HHS did not refute or
otherwise engage with this data. The FY 2013 response, which more or less repeated the FY
2012 response, 77 Fed. Reg. at 53,694, rests on even shakier ground: by the time it issued this
response, HHS had had an entire year (at least) to analyze the proposal. If HHS still needed
more time to evaluate, it should at least have explained why. The FY 2012 and 2013 responses
“provide[] no basis upon which [the Court can] conclude that” the decision not to adopt an
estimate adjustment factor for those years “was the product of reasoned decisionmaking.” Cf.
Butte County v. Hogen, 613 F.3d 190, 195 (D.C. Cir. 2010) (citation omitted).
The Secretary’s most promising argument in support of its explanations is that the
comments and their supporting data were not significant enough to merit a response more
detailed than the one HHS gave. The Secretary insists that “[t]he comments . . . did not
demonstrate, and did not even purport to demonstrate, that the proposed ‘estimate adjustment
factor’ would operate as a reliable predictor of future differences between projected and actual
payments.” Def.’s Summ. J. Mem. at 29. But read in context, the comments at the very least
rely on a strongly implied premise that deriving the estimate adjustment factor from past years’
shortfalls would create a factor in line with the model’s past trend of failure, and therefore would
reasonably anticipate the level of failure the model would produce when applied to the coming
year. AR (FY 2013) at 13100–01 (comment submitted for FY 2013); 76 Fed. Reg. at 51,793
(“The commenter stated that this would fulfill the statutory requirement in section 1886

51

(d)(5)(A) of the Act that requires that CMS establish thresholds such that outlier payments will
be projected to achieve at least 5.1 percent of DRG payments and would more closely achieve a
result that is fully consistent with the statute.”).
Similarly, the Secretary is simply wrong that the comments were bereft of “evidence or
analysis supporting the use of an ‘estimate adjustment factor.’” Def.’s Summ. J. Mem. at 29.
The comments presented an analysis and table purporting to show that “actual outlier payments
have averaged more than 0.5 percentage points below the 5.1% target of actual total DRG
payments” in most years since 2003 (FY 2013 comment) or 2004 (FY 2012 comment). AR (FY
2013) at 13100–01, 13,157; 76 Fed. Reg. at 51,793.14 They also provided concrete examples to
demonstrate how the proposed estimate adjustment factors for each year matched recent
underpayments. AR (FY 2013) at 13101 (“For example, the average shortfall from FYs 2009
and 2011 was 0.68 percentage points based on the VHDC studies. The threshold should be
modeled to produce a payment level of 5.78 percent for FY 2013.”); 76 Fed. Reg. at 51,793
(“The commenter provided an example and computed this factor for FY 2009 and FY 2010 by
taking the average variance in the actual payment for FY 2008 and FY 2009 which was 0.491
percent. Based on this factor, CMS would model the threshold to a level of 5.591 percent (5.1
plus .491 percent).”).
Thus, the FY 2012 and 2013 comments were considerably more developed in their data
support and plausibility than comments which courts have held not significant enough to require
an in-depth response. See Thompson, 741 F.2d at 409 (“None of the comments singled out by
appellant as raising substantial issues contained any meaningful analysis or data refuting the

The Court does not mean to suggest that HHS was required to accept the commenters’
analysis of this data, but the presentation was plausible enough that HHS should at least have
engaged with it.
14

52

agency’s conclusions.”); Am. Great Lakes Ports Ass’n v. Zukunft, 296 F. Supp. 3d 27, 53–54
(D.D.C. 2017) (holding that the an agency’s failure to respond to a commenter suggestion that it
include a “truing up” mechanism to compensate rate payers who had overpaid due to “disparities
between projections and actual data” did not violate the APA because the suggestion was
presented only obliquely and in one sentence of a forty-page comment) (citation omitted)), aff’d,
962 F.3d 510 (D.C. Cir. 2020). If HHS disagreed with the commenter’s premise and/or the
evidence supporting it, it should have explained why. To the extent that the Secretary now seeks
to raise concerns about the method and data that he did not even hint at in the FY 2012 and 2013
rules—such as his suggestion that the adjustment factor would have “introduc[ed] a new source
of potential distortion,” Def.’s Reply Mem. Supp. Cross Mot. Summ. J. at 19, ECF No. 196—
these are “impermissible post hoc rationalizations and thus are not properly before” the Court,
Dep’t of Homeland Sec., 140 S. Ct .at 1909. “In sum,” the estimate adjustment factor
“alternative was neither frivolous nor out of bounds and the [the agency] therefore had an
obligation to consider it.” Chamber of Com. of U.S. v. SEC, 412 F.3d 133, 145 (D.C. Cir. 2005).
The FY 2012 and 2013 responses do not reflect reasoned consideration of this alternative.
FY 2014. Finally, the Court holds that the FY 2014 rule was not arbitrary and capricious
insofar as it did not adopt or discuss the suggested estimate adjustment factor (which the
commenter renewed during the FY 2014 comment period). At first, HHS’s omission of any
discussion of the estimate-adjustment-factor proposal, which it had for two years promised to
study, seems like a shocking oversight that makes the FY 2014 response even more arbitrary and
capricious than the FY 2012 and FY 2013 responses. But a closer look at the comment in
context reveals that HHS did in effect respond to it.

53

As presented during the FY 2014 rulemaking, the estimate-adjustment-factor suggestion
was just one part of a lengthy comment that suggested several alternative methods for improving
the accuracy of HHS’s fixed loss threshold projections. AR (FY 2014) at 015852–65. The
commenter proposed adoption of an estimate adjustment factor only as a fallback alternative to
several preferred methods. The commenter wrote that it “would prefer that CMS implement the
other improvements included in our comments. However, if the other improvements are not
implemented we believe application of this adjustment would be necessary to accurately project
and pay outlier payments at the 5.1 percent target.” Id. at 015865. The “other improvements”
the commenter suggested included a proposal for calculating “actual outlier payment percentages
based on actual historical payment data,” id. at 015853 (cleaned up), and employing a projection
method a consultant had developed, id. at 015859. The consultant’s principal method involved
changing the projection period for CCRs from the one-year period HHS had been using and
estimating the fixed loss threshold using the most recent CCRs from provider specific files. Id.
at 015859–015863. The consultant also presented “two alternate methods” that “yield[ed] results
very close” to those of the consultant’s principal method: projecting CCRs quarter-by-quarter or
“estimating the rate of change in CCRs” by using “a recent historical industry-wide average as
the projection factor,” the “exact[] approach [HHS] uses to project charge inflation.” Id. at
015863–64.
In the FY 2014 final rule, HHS finally abandoned its complicated method of determining
the CCR adjustment factor by comparing the operating cost and market basket changes (the
method discussed at length in the Court’s preceding procedural analysis) in favor of the simpler
method of “us[ing] historical data to adjust the CCRs.” 78 Fed. Reg. at 50,978. Specifically,
HHS projected CCRs by “compar[ing] the percentage change in the national average case-

54

weighted operating CCR and capital CCR from the December 2011 update of the PSF to the
national average case-weighted operating CCR and capital CCR from the December 2012 update
of the PSF”—a method “consistent with [its] estimation of charge inflation.” Id. at 50,978–79.
In other words, HHS adopted the last of the commenter’s proposals for improving the fixed loss
threshold projection, an alternative method that “result[ed] in a close approximation” to the
commenter’s consultant’s principal method. AR (FY 2014) at 015864. HHS’s adoption of one
of the commenter’s suggested accuracy-improvement options sufficed to respond to the
commenter’s estimate-adjustment method, which it proposed only as a fallback in the event HHS
rejected its other preferred options. See id. at 015865 (“[I]f the other improvements are not
implemented we believe application of [an estimate adjustment factor] would be necessary to
accurately project and pay outlier payments at the 5.1 percent target.” (emphasis added)).
To be sure, in an earlier sentence, the commenter said that HHS should employ an
estimate adjustment factor if it did “not adopt all the suggestions mentioned in our comments to
improve the outlier projections.” Id. (emphasis added). HHS did not adopt all of the
commenter’s myriad suggestions. But this commenter statement was somewhat confusing,
because some of its suggestions were presented as alternatives—it would not necessarily have
made sense for HHS to adopt them all. And the change in the CCR adjustment-factor derivation
was enough of a shift in the way HHS projected fixed loss thresholds that it was reasonable for
HHS to rest upon this change as its response to the commenter’s multi-pronged comment
directed toward improving projection accuracy. Plaintiffs have not engaged with the conditional,
fallback nature of the commenter’s estimate-adjustment-factor suggestion. See Pls.’ Summ. J.
Mem. at 34. Under these circumstances, the Court declines to hold that the FY 2014 rule was
arbitrary and capricious insofar as it did not discuss the estimate adjustment factor.

55

3. HHS’s explanation for refusing to account for reconciliation in the FY 2013 rule was arbitrary
and capricious, but HHS sufficiently explained this decision in the FY 2014–2016 rules.
Plaintiffs next claim that the FY 2013–2016 rules were arbitrary and capricious because
HHS refused to account for the effects of its payment reconciliation process on its projections of
the fixed loss threshold. Pls.’ Summ. J. Mem. at 36. A bit of background is necessary to
understand this set of challenges, for the Court does not write on a clean slate with respect to
these claims. First, a refresher on HHS’s reconciliation process, which HHS has used since 2003
to fight the practice of turbo-charging: HHS can “recalculate a hospital’s eligibility [for an
outlier payment] using actual cost data at the time of settlement,” as opposed to the older cost
data used to determine the hospital’s initial CCR, and, accordingly, its eligibility for a particular
outlier payment. Billings Clinic, 901 F.3d at 306. If it turns out that the hospital was not in fact
eligible for the outlier payment it received, HHS can “claw-back” the undue amount via the
reconciliation process. Id. Thus, by recovering sums through reconciliation, HHS might end up
ultimately paying less in outlier payments for a given year than initial projections suggested.
Over the years, “commenters [have] repeatedly argued” that HHS should incorporate an
expectation that some payments would be reclaimed through reconciliation into its process for
projecting what fixed loss threshold is appropriate to achieve outlier payments that are 5.1
percent of total payments. Pls.’ Summ. J. Mem. at 35. But HHS repeatedly declined to do so.
In Banner Health, hospitals argued that it was arbitrary and capricious in the FY 2007
rule for HHS not to account for payments that would be reclaimed during reconciliation. 867
F.3d at 1356. HHS had explained its decision not to incorporate reconciliations in part on the
ground that it had implemented policies it believed would put an end to turbo-charging, and
therefore to the need for significant resort to the reconciliation process: “[W]e . . . continue to
believe that, due to the policy implemented in the June 9, 2003 outlier final rule, CCRs will no

56

longer fluctuate significantly and, therefore, few hospitals will actually have these ratios
reconciled upon cost report settlement.” 71 Fed. Reg. at 48,149. HHS further explained that
reconciliation recoupments would be difficult to predict and that its projection model aimed to
account for accurate CCRs, so should reasonably estimate post-reconciliation payments anyway:
In addition, it is difficult to predict which specific hospitals will have CCRs and
outlier payments reconciled in their cost reports in any given year. We also noted
that reconciliation occurs because hospitals’ actual CCRs for the cost reporting
period are different than the interim CCRs used to calculate outlier payments when
a bill is processed. Our simulations assume that CCRs accurately measure hospital
costs and, therefore, are more indicative of postreconciliation than prereconciliation outlier payments.
Id. In Banner Health, hospitals argued that even in spite of the anti turbo-charging reforms it
was arbitrary and capricious for the FY 2005 rule “not to forecast that particular hospitals would
continue collecting outlier payments significantly higher than their actual costs.” 867 F.3d at
1352. Concluding that HHS’s explanation that the 2003 reforms would effectively eliminate
turbo-charging was reasonable, the D.C. Circuit held that the FY 2005 rule was not arbitrary and
capricious. Id. The Court then incorporated this same analysis to reject the hospitals’ similar
challenge to the FY 2007 rule. Id. at 1356.
In FYs 2008–2011, HHS refused to take reconciliation into account, each time giving
essentially the same explanation it had given in the FY 2007 final rule. Pls.’ Summ. Jr. Mem. at
35–36 & 36 n.20 (collecting sources). A group of hospitals again sought to challenge these
refusals in Billings Clinic. As relevant here, the hospitals argued that HHS’s assertions that the
2003 reforms obviated the need for reconciliation were undercut by data showing that
reconciliations continued in spite of the reforms. They further argued that HHS’s claim of
difficulty in predicting which hospitals would be subject to reconciliation was not a good reason
to fail to take account of their important impact and that in any event, what mattered was not

57

which specific hospitals would engage in reconciliation but rather the overall total of reconciled
payments. Final Brief for the Appellants at 38–43, Billings Clinic, 901 F.3d 301 (No. 17-5006).
The D.C. Circuit rejected these challenges in Billings Clinic. It held that the Banner
Health court’s conclusion that HHS “‘was under no obligation’ to ‘account for the possibility of
reconciliation in setting the fixed-loss threshold’ . . . applie[d] with equal force to the 2008
through 2011 outlier thresholds.” Billings Clinic, 901 F.3d at 313 (quoting Banner Health, 867
F.3d at 1356). Nothing in the Billings Clinic record, the court held, supported a departure from
the holding that HHS “reasonably concluded ‘that [the] charging practices would not fluctuate
significantly enough to justify accounting for reconciliation[.]’” Id. (quoting Banner Health, 867
F.3d at 1352).
Accordingly, it is the law of the D.C. Circuit that HHS’s decisions not to account for
reconciliation in the FY 2007–2011 rules, based on the explanations repeated across those
rulemakings, were not arbitrary and capricious even in the face of the arguments hospitals made
in Billings Clinic. In refusing to account for reconciliation in the FY 2013–2016 rules, as
relevant here, HHS both repeated the explanations it had used since 2007 (that the 2003 reforms
had reduced the volume of reconciled payments, that there was difficulty in predicting which
hospitals would be subject to reconciliation, and that the model already used accurate CCRs),
and added additional justifications in response to specific comments received for FYs 2014–2016
(discussed below). 77 Fed. Reg. at 53,695 (FY 2013 Rule); 78 Fed. Reg. at 50,979–80 (FY 2014
Rule); 79 Fed. Reg. 49,854. 50,375–78 (Aug. 22, 2014) (FY 2015 Rule); Fed. Reg.49,426,
49,781 (Aug. 17, 2015) (FY 2016 Rule). To succeed on their claims that it was arbitrary and
capricious for HHS not to account for reconciliation in the FY 2013–2016 rules, Plaintiffs must
demonstrate some difference between their challenges and the ones rejected in Billings Clinic.

58

They must point to evidence in the records for the FY 2013–2016 rules that casts new doubt
upon HHS’s previous explanations (repeated in support of the FY 2013–2016 rules), and/or show
that HHS’s new justifications were arbitrary. See Billings Clinic, 901 F.3d at 313 (holding that
“[n]othing in the current record” supported a departure from Banner Health’s decision to uphold
the 2005 and 2007 refusals). Plaintiffs have done so for the FY 2013 rule, but not for the others.
Plaintiffs’ general reconciliation arguments. Plaintiffs take HHS’s repeated justifications
one at a time. Their first attack, on HHS’s explanation that “it is difficult to predict the specific
hospitals that will have CCRs and outlier payments reconciled in any given year,” e.g., 77 Fed.
Reg. at 53,695, comes on two fronts. First, Plaintiffs argue that it is “irrational” to focus on
identification of specific hospitals when what matters is “what the overall total amount of
reconciliation will be.” Pls.’ Summ. J. Mem. at 37. But the Billings Clinic challengers raised
this exact argument in challenging the FY 2008–2011 rules, and the D.C. Circuit rejected it
(albeit without specificity): this argument was part of the “record” the Billings Clinic court held
did not “support[] a different answer” than Banner Health’s decision to uphold the FY 2007 rule,
Billings Clinic, 901 F.3d at 313, which had included the same HHS explanation, 71 Fed. Reg. at
48,149. See Final Brief for the Appellants at 42, Billings Clinic, 901 F.3d 301 (No. 17-5006)
(“HHS’s justification is not logical. To account for reconciliation in setting the threshold, what
mattered was the total amount of reconciled payments, not the identities of the hospitals with
payments at issue. That it was difficult to say which hospitals might undergo reconciliation does
not imply that HHS could not estimate their impact.”). Plaintiffs have not pointed to any
differences in the rulemaking records for the FY 2013–2016 rules that would make this argument
more persuasive in opposition to these rules than it was in opposition to the FY 2008–2011 rules.
Moreover, they have not explained why it was irrational for HHS to be concerned about the

59

feasibility of identifying which hospitals would be subject to reconciliation as a means of
projecting the total amount of reconciled payments—a plausible reading of HHS’s explanations.
See Rural Cellular Ass’n v. FCC, 588 F.3d 1095, 1105 (D.C. Cir. 2009) (“[W]hen an agency’s
decision is primarily predictive, our role is limited; we require only that the agency acknowledge
factual uncertainties and identify the considerations it found persuasive.”).
In their second challenge to this first explanation (the difficulty in identifying hospitals
subject to reconciliation), Plaintiffs shift toward accepting it on its terms, but argue that specific
items in comments from 2013–2016 showed that HHS could in fact identify specific hospitals.
During the FY 2013–2016 rulemakings, a commenter reviewed hospital cost reports and
presented tables showing the total amount of outlier reconciliation payments for years 2003–
2010. AR (FY 2013) at 13103; AR (FY 2014) at 15546; AR (FY 2015) at 18545; AR (FY 2016)
at 20569. But Plaintiffs have not explained why these evaluations of past reconciled payments
(which do not even mention specific hospitals, though presumably are drawn from hospitalspecific data), responded to HHS’s concern about the difficulty in predicting specific hospitals
that would be subject to future reconciled payments. In FY 2013, the commenter also said,
without citation and rather generally, that it was “aware of hospitals that have received
reconciliation notices for significant amounts that are not reflected in” the table submitted for
that year. AR (FY 2013) at 13103. This vague assertion hardly pointed HHS toward a viable
method of predicting reconciliation payments.
Plaintiffs characterize HHS’s next repeated explanation as “the statement ‘that few
hospitals will actually have these amounts reconciled.’” Pls.’ Summ. J. Mem. at 36 (quoting 77
Fed. Reg. at 53,692). Zooming out a bit, it becomes clear that this statement was part of HHS’s
contention that the 2003 reforms would sufficiently diminish the need for reconciliation such that

60

reconciliation recoupments would not much affect total payment outlays: “[W]e continue to
believe that, due to the policy implemented in the June 9, 2003 outlier final rule (68 FR 34494),
CCRs will no longer fluctuate significantly and, therefore, few hospitals will actually have these
ratios reconciled upon cost report settlement.” 77 Fed. Reg. at 53,659. The D.C. Circuit upheld
this very same prediction as “not . . . arbitrar[y]” in Banner Health, 867 F.3d at 1351–52, and as
“reasonabl[e]” in Billings Clinic, 901 F.3d at 1313, so Plaintiffs must point to a clear difference
between the records at issue in those cases and the rulemaking records for the FY 2013–2016
rules challenged here in order to succeed in arguing that this explanation is “a non sequitur,”
Pls.’ Summ. J. Mem. at 36.
Their first effort to do so falls flat. Plaintiffs say that what matters is the total amount of
payments that will be recouped via reconciliation, not whether the number of hospitals that
participate in reconciliation is large or small. For support, they note that “[i]n the FY 2004 rule,
HHS identified 50 hospitals that it thought would be subject to reconciliation.” Id. at 36 (citing
68 Fed. Reg. at 45,476). Even though this was only a few hospitals, “HHS accounted for their
reconciliations,” which it explained would improve its estimate of the final amount of outlier
payments. Id. (citing 81 Fed. Reg. 3,727, 3,729 (Jan. 22, 2016) (HHS’s further explanation of
the FY 2004 rule after a remand)). Plaintiffs overlook the fact that the Billings Clinic
challengers also invoked the FY 2004 rule’s accounting for 50 reconciled hospitals, to no avail.
Final Brief for the Appellants at 41, Billings Clinic, 901 F.3d 301 (No. 17-5006) (“In 2003 HHS
overhauled the outlier regulations due to only about twice as many turbo-charging hospitals, and
in the 2004 threshold rulemaking HHS considered the reconciliation of about 50 hospitals to be
critical to the analysis. It is easy to see why it would be. Reconciliation was intended for
hospitals that ‘disproportionately benefited’ from vulnerabilities in the payment regulation so the

61

50 hospitals would likely be among the largest recipients. HHS has not . . . explained why it
departed from its prior practice and ignored reconciliation for the years here at issue.” (citations
omitted)).
Moreover, Plaintiffs have not explained why HHS’s FY 2004 conclusions from the
record before it then should be relevant to evaluating assertions based on the rulemaking records
for FY 2013–2016. Perhaps HHS during those later years lacked the hospital-specific data it had
had in 2004—data from before the recent enactment of the anti-turbo-charging reforms may have
made it easier to identify hospitals especially susceptible to reconciliation in 2004. See 68 Fed.
Reg. at 45,476 (noting that although “it is difficult to project which hospitals will be subject to
reconciliation of their outlier payments using available data,” HHS had been able to identify
about 50 “hospitals that ha[d] been consistently overpaid recently for outliers” (emphasis
added)). None of the comments submitted during the FY 2013–2016 rulemakings suggested a
viable way to identify any particular hospitals that would need to engage in a large volume of
reconciliation during those years.
Still, something important did change between the records at issue in Banner Health and
Billings Clinic and the FY 2013–2016 records at issue here: in each year, a comment presented
data that purported to show that a sizable volume of reconciliation had occurred between 2003
and 2009 and/or 2010 despite the long-running operation of the 2003 anti-turbo-charging
reforms. Generally, the comments each year posited that the average yearly total of reconciled
payments in this period was between about $12.3 million and $13.6 million. AR (FY 2013) at
13103 (showing a 2004–2009 yearly average of $13,570,975); AR (FY 2014) at 15546 (showing
a 2003–2009 yearly average of $12,256,814); AR (FY 2015) at 18545 (showing a 2003–2010
yearly average of $13,616,803); AR (FY 2016) at 20569 (showing a 2003–2010 yearly average

62

of $13,616,803). This data at least facially called into question the core premise of HHS’s longrunning explanation for not accounting for reconciliation payments when projecting the
threshold: that the 2003 reforms would largely eliminate the need for reconciliation. See, e.g., 71
Fed Reg. at 48,149 (FY 2007 rule); 77 Fed. Reg. at 53,695 (FY 2013 rule). Therefore, once
faced with this new data, it became arbitrary for HHS to simply repeat the same assertion it had
used previously without examining how it held up against the new data.
Yet that is exactly what HHS did in the FY 2013 rule. It noted the total volume of
reconciliation payments the commenter had identified for the 2003–2010 period (though not the
yearly average), but then simply rehearsed its old assertion: “[W]e continue to believe that, due
to the policy implemented in the June 9, 2003 outlier final rule (68 FR 34494), CCRs will no
longer fluctuate significantly and, therefore, few hospitals will actually have these ratios
reconciled upon cost report settlement.” 77 Fed. Reg. at 53,695. HHS’s failure to engage with
data apparently showing that reconciliation had continued long past the 2003 reforms is all the
more striking given that the commenter expressly requested that HHS disclose “data showing
that the recoveries obtained during the reconciliation process are immaterial.” AR (FY 2013) at
13103.
In this litigation, the Secretary responds by asserting that the commenters’ yearly
averages did not present a consistent pattern and indeed masked inconsistency in the data: the
commenter’s own 2013 table, for example, showed that annual outlier payments had jumped
from as low as $652,845 in 2003, to $7,317,950 in 2006, all the way up to $32,701,911 in 2004.
AR (FY 2013) at 13102. Because “the totals for each year varied widely,” the comments did not
“demonstrate that making an adjustment for reconciliation was likely to improve the accuracy of
the Secretary’s projections, rather than simply introducing a new source of potential distortion.”

63

Def.’s Summ. J. Mem. at 33–34. But these points cannot save the rule. For one thing, the
observation that reconciliations varied from year to year does not precisely defend the assertion
that reconciliations would generally decline because of the 2003 reforms. More importantly, the
Secretary’s observations about the commenter’s data are entirely post hoc, and therefore cannot
support the FY 2013 rule. HHS should have explained its misgivings on the record. Thus,
HHS’s FY 2013 failure to engage with data that purported to undermine HHS’s conclusion on
what HHS itself evidently considered an important aspect of the problem—whether
reconciliation payments would be negligible going forward because of the 2003 reforms—was
arbitrary and capricious.
A different result obtains for FYs 2014–2016, because in those rules HHS did what it had
failed to do in FY 2013: it explained why it thought the commenter’s average yearly
reconciliation amounts were immaterial to its fixed-loss threshold projections. Each year, HHS
explained that it did not believe that the commenter’s “relatively small annual amount would
have an impact on the outlier threshold because total outlier payments are approximately $4.3
billion.” 78 Fed. Reg. at 50,980 (FY 2014 rule); 79 Fed. Reg. 49,854. 50,377 (FY 2015 rule); 80
Fed. Reg. at 49,781 (FY 2016 rule referring readers to the FY 2015 response). To argue that
these responses were arbitrary, Plaintiffs point out “that $12.3 million, the low end of the
reconciliation that commenters cited, represents 0.29% of $4.3 billion.” Pls.’ Summ. J. Mem. at
39. Plaintiffs claim that it was arbitrary for HHS to consider a 0.29% impact on total outlier
payments immaterial because HHS frequently adjusted for similarly sized forecasted changes in
payments based on the CCR adjustment factor when setting the threshold. Id. (citing a
forecasted CCR decrease of 0.27 percent in FY 2007 and a 0.27 percent increase in FY 2008).

64

This is an apples and oranges comparison. HHS’s goal was to predict outlier payments as
accurately as feasible in order to project the appropriate threshold. When forecasting payment
changes based on CCR adjustment factors, HHS was using the outputs of a carefully calibrated
model it had already developed. HHS had these numbers, and felt confident in their accuracy, so
it made sense to use them, whatever their magnitude. In contrast, HHS’s assertion that 0.27
percent of total payments reconciled was not a large enough impact to adjust for must be
understood together with the further explanation that shortly followed, that it was difficult to
predict reconciliation for a future year because it was difficult to predict which hospitals would
be subject to reconciliation—an explanation that the Court has already held was reasonable. See
78 Fed. Reg. at 50,980 (FY 2014 rule); 79 Fed. Reg. 49,854. 50,377 (FY 2015 rule); 80 Fed.
Reg. at 49,781 (FY 2016 rule referring readers to the FY 2015 response). HHS’s apparent
conclusion that a 0.29 percent impact was not worth incorporating, understood alongside the
accompanying conclusion that HHS did not have a ready means of predicting this figure each
year and thought it would be difficult to develop one, is not inconsistent with its conclusions that
the 0.27 percent outputs of the carefully CCR calibrated model were worth incorporating. Cf.
Hall v. McLaughlin, 864 F.2d 868, 873 (D.C. Cir. 1989) (“[I]f the court itself finds the past
decisions to involve materially different situations, the agency’s burden of explanation about any
alleged ‘departures’ is considerably less.”). In contrast, cases on which Plaintiffs rely to argue
that HHS was required to explain this apparent inconsistency involved either direct reversals of
previous policies or other inapposite fact patterns. United Steel v. Mine Safety & Health Admin.,
925 F.3d 1279, 1284–85 (D.C. Cir. 2019) (agency shift from a policy requiring certain
examination to take place prior to a miner’s shift to a policy permitting a miner to begin working
before the examination had taken place); County of Los Angeles, 192 F.3d at 1022 (HHS had not

65

adequately explained why it had concluded that data was too unreliable to use for one purpose
but then used the same data for another purpose); Dist. Hosp. Partners, L.P., 786 F.3d at 57–60
(HHS had not sufficiently explained why its notice of proposed rulemaking identified 123 turbocharging hospitals to account for reconciliation, while its final rule identified only 50 turbocharging hospitals when making the same determination). Nor did any commenter alert HHS to
the alleged inconsistency.
The remaining HHS explanation, repeated in some form across FYs 2007–2011 and
2013–2016, related to the fact that reconciliation occurred because the later, updated, “actual
CCRs for the cost reporting period are different than the interim CCRs used to calculate outlier
payments when a bill is processed.” E.g., 77 Fed. Reg. at 53,695 (FY 2013 final rule). HHS
explained that its model sought to forecast CCRs accurately, so its model was closer to the later,
more accurate CCRs that any reconciliation proceeding would ultimately rely on to determine
the extent to which an outlier payment needed to be recouped. In other words, HHS thought its
model already corrected, at least to an extent, for the same imperfections in CCRs that the
reconciliation process sought to correct for, so there was no great need to factor reconciliation
into the model. E.g., id. Admittedly, HHS spelled out its reasoning in the FY 2007 rule more
clearly than it did in the later rules, but the Court understands HHS to have raised substantially
the same point across the FY 2007–2011 rulemakings. Compare, e.g., id.(FY 2013 final rule)
(“Our simulations assume that CCRs accurately measure hospital costs based on information
available to us at the time we set the outlier threshold.”), with 71 Fed. Reg. at 48,149 (FY 2007
final rule) (“Our simulations assume that CCRs accurately measure hospital costs and, therefore,
are more indicative of postreconciliation than pre-reconciliation outlier payments. As a result,

66

we proposed to continue to omit any assumptions about the effects of reconciliation from the
outlier threshold calculation.”).
Plaintiffs say that this explanation is “simply illogical” because HHS’s model drew upon
CCR data from old cost reports (settled before March of the preceding fiscal year), so HHS could
not assume that the CCRs in its model were accurate predicters of post-reconciliation payments,
which relied on later, updated cost reports. Pls.’ Summ. J. Mem. at 37–38. It is not clear to the
Court that Plaintiffs have demonstrated any failure of logic in HHS’s reasoning; as Plaintiffs
acknowledge, HHS did not simply take the March CCRs but employed a model to project them
forward to approximate what they would be in the coming year. See id. But the Secretary does
not defend HHS’s reasoning on this ground. Instead, he says that the accurate CCR forecast
explanation was ancillary to HHS’s “primary” concern that it would be difficult to model
reconciliation payments. Def.’s Summ. J. Mem. at 32–33. In any event, the Court holds that
HHS’s FY 2014–2016 rules find sufficient support in the non-arbitrary explanations regarding
the contextual insignificance of the amounts of reconciliation payments the commenter identified
and regarding the difficulty in identifying hospitals subject to reconciliation. They therefore
stand regardless of the logical strength of the accurate-CCR-forecast explanation.
Plaintiffs’ arguments regarding the 2013 OIG Report. The Court reaches a similar
conclusion with respect to the last of Plaintiffs’ reconciliation-related challenges. During the FY
2015 and FY 2016 comment periods, a commenter drew HHS’s attention to an OIG report from
2013 which indicated that 158 hospitals had received an usually large amount of outlier
payments over the course of 2008–2011. AR (FY 2015) at 18808–10; AR (FY 2016) at 20611–
20614. The commenter suggested that these unusually high payments would “presumably be
recouped” via reconciliation, and that HHS should factor this anticipated reconciliation recovery

67

into its projection of the fixed loss threshold. AR (FY 2015) at 18808–10; see AR (FY 2016) at
20614. Plaintiffs claim that HHS’s refusal to account for reconciliation related to the hospitals
identified in the 2013 OIG report in both the FY 2015 and FY 2016 final rules was arbitrary and
capricious. Pls.’ Summ. J. Mem. at 40; Pls.’ Combination Summ. J. Opp’n & Reply (“[I]n the
FYs 2015 and 2016 rulemakings, HHS irrationally failed to take account of the likelihood there
would be additional reconciliation, given the high-charge claims discussed in the 2013 report
from HHS’s Office of Inspector General.”); see 79 Fed. Reg. 49,854. 50,377–78 (FY 2015 rule);
80 Fed. Reg. at 49,781–82 (FY 2016 rule).
The Court begins its evaluation of this claim where it left off with Plaintiffs’ previous
reconciliation arguments. In addition to the reasons HHS gave for not relying on the OIG report,
the FY 2015 and 2016 rule’s decisions not to account for any reconciliation rested on the same
explanations the Court has just held were not arbitrary: the insignificance of the amounts of
reconciliation payments presented in comments and the difficulty in identifying which hospitals
would be subject to reconciliation. See 79 Fed. Reg. at 50,377 (FY 2015 rule); 80 Fed. Reg. at
49,781 (FY 2016 rule) (referring readers to the FY 2015 explanations). Therefore, Plaintiffs
must show that the OIG report either undermined these explanations as HHS used them in FYs
2015 and 2016 or so altered the record that HHS was required to reach a different result in spite
of these valid explanations. Plaintiffs have not done so.
For one thing, it is not as if the OIG report definitively concluded that the hospitals would
be subject to reconciliation or were engaged in improper charging practices, much less that any
such reconciliation would occur for FYs 2015 and 2016 in particular. As Plaintiffs acknowledge,
“[t]he OIG did not determine that any of the high-outlier hospitals had engaged in misconduct or
identify what recoupments might be warranted.” Pls.’ Summ. J. Mem. at 40 (citing AR (FY

68

2016) at 20663). The OIG report allowed that in at least some cases, high charges could
accurately match high costs of a treatment. See AR (2015 Rule) at 18839. Thus, the OIG report
did not by itself give rise to a compelling need to account for reconciliation in FY 2015 or FY
2016.
For another, though HHS’s responses to the OIG-report comments may not have been
uniformly airtight, Plaintiffs have not met their burden of persuading the court that they were
arbitrary and capricious. First, Plaintiffs take issue with HHS’s FY 2015 statement that any
unduly high charges would be mitigated by the application of CCRs, which would reduce outlier
payments for these hospitals to their proper level. Pls.’ Summ. J. Mem. at 40–41; 79 Fed. Reg.
at 50,377. Plaintiffs say that this assertion contradicts the very premise of the reconciliation
policy, that CCRs might be inaccurate because they might “lag certain charge increases.” Id.
But as the Secretary points out, the possible lag between CCRs and charge increases for which
reconciliation is designed to correct does not mean that a high-charging hospital’s CCR will
necessarily be incorrect, because high charges do not necessarily indicate that a hospital has
suddenly ramped up charges for non-cost-related reasons (as occurs in the turbo-charging
technique reconciliation is designed to stop). “A cost-to-charge ratio would not overstate costs,
for example, if the hospital’s charges were already high at the time the cost-to-charge ratio was
computed, or if the hospital’s charges have increased in step with parallel increases in the
hospital’s costs.” Def.’s Summ. J. Mem. at 37. Again, the OIG report did not accuse the subject
hospitals of turbo-charging or any other form of charge manipulation, and it acknowledged that
at least some of the high charges might have been legitimate. Therefore, HHS’s 2015
expectation that application of CCRs would obviate the need to subject the OIG-report hospitals
to reconciliation was not obviously inconsistent with its reconciliation policy.

69

Second, Plaintiffs argue that it was irrational for HHS to say in the FY 2015 rule that the
hospitals flagged in the OIG report would not end up reconciling because they probably did not
meet HHS’s criteria for automatic reconciliation. See 79 Fed. Reg. at 50,377–78. Plaintiffs say
that this assumption ignored HHS’s policies that a fiscal intermediary could, in its discretion,
subject any hospital to reconciliation, even if the hospital did not automatically qualify, and that
an intermediary could, if it believed a hospital was using an inaccurate CCR, “specify an
alternative CCR.” Pls.’ Summ. J. Mem. at 41 (citation omitted). The FY 2016 explanation even
expressly acknowledged the possibility of specification of alternative CCRs. 80 Fed. Reg. at
49,781; see Pls.’ Summ. J. Mem. at 41–42. But nothing in the OIG report or the associated
comment gave HHS any reason to believe that fiscal intermediaries would apply their
discretionary reconciliation or alternative CCR authority to the hospitals the OIG report flagged.
See Def.’s Summ. J. Mem. at 37.
Third, in FY 2016, the commenter questioned HHS’s FY 2015 assertion that the
application of CCRs would prevent the OIG-report hospitals from receiving undue outlier
payments by referring HHS to the OIG report’s observation that the flagged hospitals had
materially similar CCRs to other hospitals (and had similar lengths of stay)—yet still charged
substantially more for the same procedures. AR (FY 2016) at 20613. HHS responded in the FY
2016 final rule, in part, by noting that the 2008–2011 CCRs used in the OIG report may not have
been fully updated:
The CCRs are updated in the PSF at the time the MAC tentatively settles the
hospital cost report, which is approximately 6 to 7 months after the cost report has
been submitted. Thus, there is a lag in CCRs with the possibility that a CCR may
be 18 months old from the time the cost report is submitted by the provider to the
MAC until it is updated at the following tentative settlement. Because hospitals
typically increase their charges, over time CCRs will decrease but, due to the lag
these lower CCRs will not be reflected in the PSF until the following tentative

70

settlement. Thus, it is possible that the PSF will reflect CCRs that are similar for
hospitals with high and low outlier payments.
80 Fed. Reg. at 49,781. Plaintiffs claim that this response “[c]ontradict[ed] HHS’s response
from the previous year,” but they do not explain how. Pls.’ Summ. J. Mem. at 41. This response
was an explanation of HHS’s previous-year statement—that application of CCRs would mitigate
unduly high hospital charges—in response to the 2016 commenter’s challenge to this statement,
not a contradiction of it. Plaintiffs also attack HHS’s response because HHS “did not explain
why . . . high-outlier claims were irrelevant. The time lag in updating CCRs does not prevent
HHS from reconciling a hospital’s outlier payments, it is the whole point of reconciliation.” Id.
But this argument essentially re-hashes Plaintiffs’ objection to the initial 2015 CCR explanation,
which, as the Court has explained, is not clearly inconsistent with reconciliation’s purpose of
preventing hospitals from taking advantage of the time-lag in CCRs.
Fourth, and finally, Plaintiffs point to another part of HHS’s response to the FY 2016
comment. HHS observed that hospitals “determine what they will charge for items, services, and
procedures provided to patients, and these charges are the amount that the [hospitals] bill for an
item, service, or procedure. Moreover, different hospitals can have similar lengths of stay but
different CCRs.” 80 Fed. Reg. at 49,781–82. Plaintiffs complain that this statement ignored that
the OIG report compared hospitals with similar lengths of stay. Pls.’ Summ. J. Mem. at 41. To
the contrary, this statement responded to this feature of the OIG report and explained why HHS
thought the report did not undermine HHS’s conclusion that lower CCRs would reduce outlier
payments to the OIG-report hospitals. Absent any further argument from Plaintiffs on this point,
HHS’s response was not arbitrary. To wrap up, the 2013 OIG report and Plaintiffs’ criticisms of
HHS’s discussion of it do not disturb the Court’s conclusion that HHS acted lawfully when it

71

declined to account for reconciliation when setting the FY 2015 and FY 2016 fixed loss
thresholds.
4. The proper remedy is remand of the FY 2012 and FY 2013 rules for further proceedings
consistent with this opinion.
At this point, it makes sense to take stock of where the foregoing discussion leaves us.
The Court has held that none of the challenged rules were procedurally deficient. However, the
FY 2012 and FY 2013 rules were arbitrary and capricious in that HHS did not sufficiently
explain its rejection of the commenter suggestion to use an estimate adjustment factor to improve
the results of its model. The FY 2013 rule suffered from the additional defect of failing to
sufficiently explain HHS’s decision not to account for reconciliation.
Plaintiffs ask the Court to vacate these rules “and remand for recalculation of the
hospitals’ outlier payments.” Pls.’ Summ. J. Mem. at 42. The Secretary contends that the Court
should instead simply remand to the Secretary for further action consistent with this opinion.
Def.’s Summ. J. Mem. at 39; Def.’s Reply Mem. Supp Cross Mot. Summ. J. at 23. Based on the
nature of the infirmities the Court has found in the FY 2012 and FY 2013 rules, the Court agrees
with the Secretary that remand is the proper course.
It is true that “[t]he ordinary practice . . . is to vacate unlawful agency action.” Standing
Rock Sioux Tribe v. United States Army Corps of Engineers, 985 F.3d 1032, 1050 (D.C. Cir.
2021). But the nature of the agency’s legal violation informs the question of the proper remedy.
When, as here, the agency has failed to sufficiently explain its rules, “the proper course, except
in rare circumstances, is to remand to the agency for additional investigation or explanation.”
Dist. Hosp. Partners, L.P., 786 F.3d at 60 (citation omitted). Before vacatur, “[i]n such
circumstances, the agency must first be afforded an opportunity to articulate, if possible, a better
explanation.” Banner Health, 867 F.3d at 1357 (citation omitted and cleaned up). Thus, when

72

the D.C. Circuit held in Banner Health that the FY 2004, 2005, and 2006 fixed-loss threshold
rules were arbitrary and capricious due to certain failures of explanation, it remanded the rules so
that HHS would have a chance “to remedy the explanatory deficiencies. Id. at 1343 , 1345,
1349–50, 1356–7; see also Dist. Hosp. Partners, LP, 786 F.3d at 63 (holding that “the 2004
outlier threshold . . . [was] inadequately explained and” instructing the district court to “remand
the 2004 rule to the Secretary for further proceedings consistent with this opinion”); County of
Los Angeles, 192 F.3d at 1023 (“While we have identified significant inconsistencies and gaps in
the Secretary’s rationale for using the 1981 MEDPAR file [to set outlier thresholds for fiscal
years 1985–1986], bedrock principles of administrative law preclude us from declaring
definitively that her decision was arbitrary and capricious without first affording her an
opportunity to articulate, if possible, a better explanation.”).
The Court follows Banner Health, District Hospital Partners, and County of Los Angeles
because they are directly on point: each remanded without vacating when it found that annual
fixed-loss threshold rules were arbitrary and capricious because they suffered from explanatory
failures. For good measure, the Court notes that the Allied Signal factors, which the D.C. Circuit
has applied to guide the vacatur-or-remand decision in many cases but not in fixed-loss threshold
cases, support the decision to remand. Under this test, “[t]o determine whether to remand
without vacatur, [courts] consider[] first, the seriousness of the action’s deficiencies, and,
second, the likely disruptive consequences of vacatur.” Am. Great Lakes Ports Ass’n v. Schultz,
962 F.3d 510, 518–19 (D.C. Cir. 2020) (cleaned up) (citing Allied-Signal, Inc. v. Nuclear Regul.
Comm’n, 988 F.2d 146, 150–51 (D.C. Cir. 1993)).
Plaintiffs offer only two reasons in support of their argument that the seriousness-ofdeficiencies factor supports vacatur. This factor “is determined at least in part by whether there

73

is ‘a significant possibility that the [agency] may find an adequate explanation for its actions’ on
remand.” Standing Rock Sioux Tribe, 985 F.3d at 1051. First, Plaintiffs say that HHS’s shift to a
simplified CCR-adjustment-factor method in FY 2014 would preclude HHS from defending its
previous method on remand. Pls.’ Summ. J. Mem. at 43. But the first specific violation the
Court has identified is not HHS’s decision to employ its complex, pre-2014 method in the FY
2012 and 2013 rules. Rather, the Court has held that HHS did not provide a sufficient
explanation during these years of its rejection of the proposal to use an estimate adjustment
factor alongside, and as a corrective to, the complex method. HHS’s adoption of a simplified
method for FY 2014 does not necessarily preclude it from coming up with an adequate
explanation for its decision not to add an estimate adjustment factor to the complex method it
used in FY 2012 and FY 2013. Plaintiffs do not argue that HHS’s FY 2014 switch to a
simplified CCR-adjustment method precludes HHS from adequately supplementing on remand
the second explanation the Court has found lacking, regarding HHS’s decision not to account for
reconciliation in the FY 2013 rule. Indeed, the decision whether to account for reconciliation
does not appear to turn on HHS’s choice of CCR-adjustment method—Plaintiffs have challenged
HHS’s decisions not to account for reconciliation both before and after the FY 2014 switch to a
simplified method.
Next, Plaintiffs say that because HHS eventually determined it could (and would)
account for reconciliation in FY 2020, HHS could not conceivably explain on remand its
decision not to do so in FY 2013. See id. But there may have been circumstances in 2013 that
made it more difficult to predict reconciliation at that time than it was in 2020. Moreover, the
specific defect the Court has identified was HHS’s FY 2013 failure to engage with commenter

74

data tending to show that a certain amount of reconciliation was taking place. HHS may be able
to sufficiently support its decision with further explanation, as it did in the FY 2014–2016 rules.
As for the consequences of vacatur, Plaintiffs identify the following: “[i]f HHS does, as
the Hospitals predict, end up lowering the threshold, hospitals that have available procedural
mechanisms may be able to seek additional outlier payments.” Id. at 44. Plaintiffs say that this
would not be disruptive because it would not require broader corrective action beyond
recalculating specific hospital cost reports. See id. But the D.C. Circuit has held that remand
without vacatur “is appropriate when vacatur would disrupt settled transactions.” Am. Great
Lakes Ports Ass’n, 962 F.3d at 519 (upholding the district court’s decision to remand without
vacating when vacating the challenged rate-setting rule would mean that past payments were
erroneous). Because any necessary adjustment of outlier payments to Plaintiffs for FYs 2012
and 2013 could wait until after the Secretary has had a chance to decide whether to revise its
responses on remand or proceed otherwise, the Court holds that this factor weighs in favor of
remanding. Therefore, the Court remands the FY 2012 and FY 2013 rules, without vacatur, for
further proceedings consistent with this opinion.
Given this disposition, the Court finds it unnecessary at this time to reach Plaintiffs’
remaining claim, that it was arbitrary and capricious in the FY 2012 and FY 2013 rules to
continue using the same CCR adjustment method even in the face of evidence that, according to
Plaintiffs, demonstrated that HHS had routinely underestimated the decline in CCRs and that it
had frequently failed to achieve its 5.1 percent payment target. See Pls.’ Summ. J. Mem. at 26–
34. Though the parties dispute the strength of the evidence supporting these asserted trends, see
Def.’s Summ. J. Mem. at 25–27, Plaintiffs’ argument on this claim is closely related to the trend
data the Court has found concerning enough to require a persuasive HHS response in the context

75

of the estimate adjustment factor comments. See supra at 50. Thus, the Court’s remand
regarding the estimate adjustment factor will require HHS to reconsider its fixed-loss threshold
projection for FYs 2012 to 2013, including in a way that relates closely to Plaintiffs’ remaining
challenge. Or, on remand, the Secretary may choose to apply a new approach to the FY 2012
and 2013 thresholds rather than defend its original method. See County of Los Angeles, 192 F.3d
at 1023 (noting that on remand, the Secretary could “either . . . recalculate outlier thresholds for
fiscal years 1985–1986 or . . . offer a reasonable explanation for refusing to use [certain data] in
setting outlier thresholds during those years”).
Because it is unclear whether Plaintiffs’ contrary-to-trend argument will take the same
shape, or be relevant at all, after remand, the Court declines to adjudicate it now. See Pub.
Citizen v. Fed. Motor Carrier Safety Admin., 374 F.3d 1209, 1216 (D.C. Cir. 2004) (holding that
a rule was arbitrary and capricious in one respect but not reaching other claims that “also raise[d]
troubling concerns about the decisionmaking process” because on remand “the agency [would]
be free in its further proceedings to consider the other objections anew in light of this opinion
and its own responses” to the issue that was the basis for the remand); cf. BP Energy Co. v.
FERC, 828 F.3d 959, 969 (D.C. Cir. 2016); Nassar & Co. v. SEC, 566 F.2d 790, 794 (D.C. Cir.
1977); Exxon Co., U.S.A. v. FERC, 182 F.3d 30, 45 (D.C. Cir. 1999); Oceana, Inc. v. Ross, 363
F. Supp. 3d 67, 70 (D.D.C. 2019). If HHS does choose to maintain its original course on
remand, it would do well to seriously engage with Plaintiffs’ complaints about the underpayment
trends they claim to have identified—Billings Clinic, 901 F.3d at 315 provides support for the
general proposition that HHS was not entitled to rely on the same method and explanations in the
face of any sustained pattern of underpayments.

76

5. The Court will not revisit its previous holding that certain of Plaintiffs’ claims are timely.
The Court’s decision to remand on the ground that the FY 2012 and FY 2013 rules were
arbitrary and capricious in the ways the Court has described requires it to resolve one additional
issue. The Secretary asks the Court to reconsider its earlier ruling that certain plaintiffs’ claims
relating to (among other rules) the FY 2012 and FY 2013 rules were timely even though they
were added to the action by amended complaint rather than supplemental complaint. Under 42
U.S.C. § 1395oo(f)(1), a party must bring any district court action pertaining to a PRRB judicialreview grant within 60 days of the decision. The plaintiffs added the claims in question within
60 days by amended complaint. Because a party may not add new claims based on events that
took place after the initial filing of the suit by amended complaint as of right, and instead must
do so by obtaining leave of court to file a supplemental complaint, the Secretary argues that these
claims were not filed within the 60-day period. As the Court summarized previously:
The Secretary suggests that, because these particular claims had been approved for
expedited judicial review after the filing of the original complaints, they should
have been added by supplementation under Rule 15(d) (which requires court
approval). As a result, according to the Secretary, these claims were filed “without
legal effect” and are now barred, because they were not actually submitted within
60-day period for filing a district court action under § 1395oo(f)(1).
Mot. Dismiss Op. at 26–27 (citations omitted). In response to the Secretary’s motion to dismiss
on this ground, the Court agreed that plaintiffs were required to add the in-question claims by
supplemental complaint rather than amended complaint, but nevertheless “exercise[d] its
discretion to treat the mislabeled amended complaint as a supplemental complaint, nunc pro
tunc.” Id. at 30. The Court explained that the Secretary had not identified any prejudice due to
the hospitals’ mislabeling of what was in effect a supplemental complaint as an amended
complaint, that the Secretary could have moved to strike the filing but instead waited until after
the 60-day period had passed before bringing the mislabeling issue to the Court’s attention, and

77

that there was no reason to think the Court would not have granted leave to file a supplemental
complaint if the hospitals had requested it at the proper time. Id. at 29–30.
Now, the Secretary asks the Court to “reconsider[]” this decision. Def.’s Suppl. Mem. at
41. Reconsideration of an interlocutory order may be appropriate “when a court has ‘patently
misunderstood the parties, made a decision beyond the adversarial issues presented, made an
error in failing to consider controlling decisions or data, or where a controlling or significant
change in the law has occurred.’” Ali v. Carnegie Inst. of Wash., 309 F.R.D. 77, 80 (D.D.C.
2015) (quoting U.S. ex rel. Westrick v. Second Chance Body Armor, Inc., 893 F. Supp. 2d 258,
268 (D.D.C. 2012)). A court’s failure to consider “controlling decisions . . . that might
reasonably be expected to alter [its] conclusion” may also be grounds for reconsideration.
Jordan v. U.S. Dep’t of Lab., 308 F. Supp. 3d 24, 35 (D.D.C. 2018) (citation omitted).
The Secretary’s principal contention is that the Court’s decision was contrary to Supreme
Court precedent holding “that a court cannot issue a nunc pro tunc order to revise the date of a
court action or otherwise alter the history of a case.” Def.’s Summ. J. Mem. at 41–42. For this,
the Secretary relies on Roman Cath. Archdiocese v. Acevedo Feliciano, 140 S. Ct. 696, 700–01
(2020) (per curiam). But the Secretary reads this case too broadly. In Acevedo Feliciano, a case
had been removed from the Puerto Rico Court of First Instance to federal district court. Id. at
700. The district court granted a motion to remand on August 20, 2018, but attempted to do so
nunc pro tunc on to May 13, 2018, the day it had likely lost subject matter jurisdiction on
related-to-bankruptcy grounds due to a bankruptcy court’s dismissal of a related bankruptcy
proceeding. Id. at 701. The Supreme Court held that orders the Court of First Instance issued
between May 13 and August 20 were void, because the Court of First Instance lacked any
jurisdiction until the case had been remanded. Id. at 700. It did not matter that the district court

78

had attempted to remand nunc pro tunc as of May 13. This attempted relation back was not
effective, because “nothing occurred in the [d]istrict [c]ourt on March 23, 2018.” Id. at 701. The
Supreme Court explained that a nunc pro tunc order may be issued only “to reflect the reality of
what has already occurred. . . . Such a decree presupposes a decree allowed, or ordered, not
entered, through inadvertence of the court.” Id. at 700–01. The purpose of a nunc pro tunc
decree “is to correct mistakes or omissions in the record so that the record properly reflects the
events that actually took place.” Rohe v. Wells Fargo Bank, N.A., 988 F.3d 1256, 1261 n.6 (11th
Cir. 2021) (citation and emphasis omitted).
Unlike in Acevedo Feliciano, in which the district court sought to relate its nunc pro tunc
decree to an event (the bankruptcy court dismissal) that occurred in another court, in this case,
the Court’s nunc pro tunc granting of leave to file a supplemental complaint did relate to
something that had occurred in the district court: the filing of the amended complaints that
should have been labeled as supplemental complaints. As the Court explained when it resolved
the motion to dismiss, to the extent the “amended complaints” added claims based on PRRB
expedited review grants that took place after the initial filing of the action, they were, by
definition, not actually “amended complaints”—despite being labeled that way. Instead, they
were properly understood as supplemental complaints. Mot. Dismiss. Op. at 27 (“The difference
between supplementation and amendment is that an amendment ‘typically rest[s] on matters in
place prior to the filing of the original pleading,’ while a supplement ‘sets forth transactions or
occurrences or events which have happened since the date of the pleading sought to be
supplemented.” (quoting United States v. Hicks, 283 F.3d 380, 385 (D.C. Cir. 2002) (internal
quotation marks omitted)). The “amended complaints” were simply “mislabel[ed].” Id. at 30.
Thus, by granting leave to add these claims nunc pro tunc, the Court was simply acting “to

79

reflect the reality of what ha[d] already occurred,” Acevedo Feliciano, 140 S. Ct. 696, 700–01
(cleaned up): the filing of a supplemental complaint. The Court did not backdate any new filing
or order; it simply “treat[ed] the mislabeled amended complaint as a supplemental complaint,
nunc pro tunc.” Mot. Dismiss Op. at 30.15 Given these differences between Acevedo Feliciano
and the instant case, Acevedo Feliciano is not a controlling decision that makes reconsideration
appropriate.16
Beyond asserted inconsistency with Supreme Court precedent, the Secretary says that the
Court’s motion-to-dismiss decision was outside the adversarial issues presented, but does not
explain how. Def.’s Summ. J. Mem. at 40. In any event, the Court does not find any of the
Secretary’s remaining arguments to justify reconsideration of its initial holding. The Secretary
takes issue with the Court’s previous conclusion that the timing of his timeliness argument had a

15

The fact that Plaintiffs did not move for leave to file a supplemental complaint when
they filed their “amended complaints” does not alter the Court’s conclusion. Had the filings
been properly labeled as a supplemental complaint, the Court would have had the discretion to
treat the supplemental complaint as including the required request to file one. See In re Regions
Morgan Keegan Sec., Derivative v. Morgan Asset Mgmt., Inc., No. 08-2260, 2010 WL
11441471, at *1 (W.D. Tenn. Jan. 4, 2010) (finding that a non-pro se plaintiff’s filing, though
labeled “Amended Complaint,” was actually a supplemental complaint and construing it as a
motion to file a supplemental complaint); cf. Mangwiro v. Napolitano, 939 F. Supp. 2d 639, 647
(N.D. Tex. 2013) (construing a supplemental complaint “as a motion to supplement Plaintiff’s
Complaint because, although not styled as a motion, it includes a request by Plaintiffs to
supplement their pleadings”), aff’d sub nom. Mangwiro v. Johnson, 554 F. App’x 255 (5th Cir.
2014); Costa v. Bazron, No. CV 19-3185, 2020 WL 1935524, at *1 (D.D.C. Apr. 22, 2020)
(“Although Plaintiffs seek leave to amend their complaint under Rule 15(a)(2), the Court will
construe the pleading as a motion to file a supplemental complaint pursuant to Rule 15(d).”
(citation omitted)). As the Court explained when denying the motion to dismiss, “there is no
reason to think the Court would not have granted the motion to supplement, given the interests in
judicial economy,” Mot. Dismiss Op. at 29–30, and even now the Secretary has not offered any
reason why such a motion should not have been granted.
16

The Secretary also cites a Supreme Court case he says indicates that the § 1395oo(f)(1
60-day period is not subject to equitable tolling, Def.’s Summ. J. Mem. at 42 (quoting Sebelius v.
Auburn Reg’l Med. Ctr., 568 U.S. 145, 159–60 (2013), but the Court’s decision that the
supplementary claims are not untimely was not and is not based on equitable tolling.

80

“‘gotcha’ quality,” Mot. Dismiss Op. at 29, insisting that he warned Plaintiffs’ counsel about the
interaction between the requirement of court leave to file a supplemental complaint and the 60day § 1395oo(f)(1) period during the Banner Health litigation. But the Court will not rest its
decision on events that took place outside the record of this case. And in any case, the Court’s
initial point remains: the Secretary could have raised this issue earlier by moving to strike the
improper amended complaint. See Mot. Dismiss Op. at 29. Moreover, the Court notes that the
Secretary’s summary judgment brief does not call into question a key premise of the Court’s
motion-to-dismiss ruling, that “the Secretary has not pointed to any undue prejudice that has
resulted from the mislabeling.” Id. at 30. The Court holds that reconsideration of is motion-todismiss holding is not warranted. Therefore, the supplemental claims are timely.
VI. CONCLUSION
For the foregoing reasons, Plaintiffs’ Motion to Complete Administrative Records (ECF
No. 184) is GRANTED IN PART AND DENIED IN PART; Plaintiffs’ Motion for Summary
Judgment (ECF No. 185) is GRANTED IN PART AND DENIED IN PART; and Defendant’s
Cross-Motion for Summary Judgment (ECF No. 188) is GRANTED IN PART AND DENIED
IN PART. An order consistent with this Memorandum Opinion is separately and
contemporaneously issued.
Dated: 06/17/2022

RUDOLPH CONTRERAS
United States District Judge

81