Banner Health v. Sebelius , 126 F. Supp. 3d 28 ( 2015 )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
BANNER HEALTH f/b/o BANNER GOOD
SAMARITAN MEDICAL CENTER, et al.,
Plaintiffs
v.
Civil Action No. 10-1638 (CKK)
SYLVIA M. BURWELL, Secretary of the
U.S. Department of Health and Human
Services,
Defendant
MEMORANDUM OPINION
(September 2, 2015)
Plaintiffs are twenty-nine organizations that own or operate hospitals participating in the
Medicare program. They have sued the Secretary of the Department of Health and Human
Services (the “Secretary”), purporting to challenge various actions taken by the Secretary in the
course of administering Medicare’s “outlier” payment system, the system which provides
additional payments to hospitals for extremely high cost cases. Plaintiffs challenge a series of
regulations that, together, govern outlier payments for federal fiscal year (“FY”) 1997 through
FY 2007. Specifically, they challenge 14 regulations: outlier payment regulations promulgated in
1988, 1994 and 2003, and 11 annual fixed loss threshold regulations issued for FY 1997 through
FY 2007. 1 As explained in more detail below, the combination of the applicable outlier payment
1
Plaintiffs’ claims in this action have long been wide ranging, and the Court previously
dismissed those claims that exceed the scope of the proper judicial review of agency action. See
Banner Health v. Sebelius, 797 F. Supp. 2d 97, 110-11 (D.D.C. 2011). Over the course of the
lengthy procedural history of this case, it has become clear that Plaintiffs in fact challenge 14
regulations, promulgated between 1988 and 2007. Notwithstanding this clarification and
narrowing of this action, Plaintiffs continue to use broad language addressed at flaws of the
outlier program overall, in addition to identifying certain purported legal flaws with the
regulations challenged. Nonetheless, in addition to addressing Plaintiffs’ claims that are directed
at specific regulations, where the connections between Plaintiffs’ other arguments and individual
1
regulations and the applicable annual fixed loss threshold establishes the formula for calculating
the outlier payments made to individual hospitals, including payments to Plaintiffs and to the
facilities controlled by Plaintiffs, during each fiscal year. Plaintiffs challenge the individual
outlier payments made by applying those 14 regulations, as well. 2
Presently before the Court are Defendant’s [126] Motion to Dismiss for Lack of Subject
Matter Jurisdiction and for Summary Judgment, Plaintiffs’ [127/142] Motion for Summary
Judgment, and Plaintiffs’ [128] Motion (Related to Their Motion For Summary Judgment) for
Judicial Notice and/or for Extra-Record Consideration of Documents and Other Related Relief.
Upon consideration of the pleadings, 3 the relevant legal authorities, and the record as a whole,
rulemakings are apparent, the Court addresses those arguments in this Memorandum Opinion as
well.
2
Plaintiffs do not claim that the agency applied the regulations incorrectly in calculating the
individual outlier payments; they claim that, because of flaws with the underlying regulations,
the resultant outlier payments were necessarily incorrect.
3
The Court’s consideration has focused on the following documents:
• Def.’s Mot. to Dismiss for Lack of Subject Matter Jurisdiction and for Summary
Judgment (“Def.’s Mot.”), ECF No. 126;
• Pls.’ Corrected Opp’n to Def.’s Mem. in Support of Mot. to Dismiss for Lack of
Subject Matter Jurisdiction and for Summary Judgment (“Pls.’ Opp’n”), ECF No.
143;
• Def.’s Reply Mem. in Support of Mot. to Dismiss for Lack of Subject Matter
Jurisdiction and for Summary Judgment (“Def.’s Reply”), ECF No. 134;
• Pls.’ Corrected Mot. for Summary Judgment (“Pls.’ Mot.”), ECF No. 142;
• Def.’s Mem. in Opp’n to Pls.’ Mot. for Summary Judgment (“Def.’s Opp’n”), ECF
No. 132;
• Pls.’ Corrected Reply in Support of Pls.’ Mot. for Summary Judgment (“Pls.’ Reply”),
ECF No. 144;
• Pls.’ Motion (Related to their Motion for Summary Judgment) for Judicial Notice
and/or for Extra-Record Consideration of Documents and Other Related Relief (“Pls.’
Extra-Record Mot.”), ECF No. 128;
• Def.’s Mem. in Opp’n to Pls.’ Mot. to Supplement Administrative Records With
Extra-Record Evidence and for Leave to Exceed Page Limit (“Def.’s Extra-Record
Opp’n”), ECF No. 129; and
2
the Court DENIES Defendant’s [126] Motion to Dismiss for Lack of Subject Matter Jurisdiction,
GRANTS IN PART and DENIES IN PART Defendant’s [126] Motion for Summary Judgment,
GRANTS IN PART and DENIES IN PART Plaintiffs’ [127/142] Motion for Summary Judgment,
and GRANTS IN PART and DENIES IN PART Plaintiffs’ [128] Motion for Judicial Notice
and/or for Extra-Record Consideration of Documents and Other Related Relief.
As an initial matter, the Court concludes that it has subject matter jurisdiction over all of
the claims in this action. With respect to the FY 2004 fixed loss threshold rule, the Court
REMANDS the rule to the agency to allow the agency to explain its decision regarding its
treatment of certain data—or to recalculate the fixed loss threshold if necessary—as explained
further below. In all other respects, the Court DENIES Plaintiffs’ challenges to all of the
regulations at issue in this case.
With respect to Plaintiffs’[128] Motion for Judicial Notice and/or Extra-Record
Consideration of Documents and Other Related Relief, the Court DENIES Plaintiffs’ request to
supplement the record and for extra-record consideration of documents, but the Court GRANTS
the motion insofar as the Court will take judicial notice of the publicly available materials
subject to the motion, as relevant. The Court DENIES Plaintiffs’ request to submit three
additional tables and STRIKES from the record exhibits 5, 7, and 8 to Plaintiffs’ Motion for
• Pls.’ Reply in Support of Mot. for Judicial Notice and/or for Extra-Record
Consideration of Documents and Other Related Relief (“Pls.’ Extra-Record Reply”),
ECF No. 130.
In light of the voluminous administrative record pertaining to numerous regulations, the Court
cites to the administrative record by referring to the specific agency action for which that portion
of the administrative record was assembled, e.g., “A.R. (FY 2004)” or “A.R. (2003
amendments).” In an exercise of its discretion, the Court finds that holding oral argument in this
action would not be of assistance in rendering a decision. See LCvR 7(f).
3
Summary Judgment. The Court will retain jurisdiction pending the limited remand to the agency
regarding the FY 2004 rulemaking.
I. BACKGROUND
A. Factual Background
While this action emerges from Plaintiffs’ challenge to the outlier payments they received
for FY 1997 through FY 2007, the hospitals do not challenge the calculation of the individual
outlier payments; instead, they level their substantive challenges at the 14 regulations that
established the formulas for outlier payments in each of the relevant years. Plaintiffs challenge
two sets of interrelated regulations that, taken together, create a formula that generates the actual
outlier payments paid each year. With these regulations in hand, the calculations of the actual
outlier payments are effectively a ministerial task. The first set of regulations consists of three
rules—promulgated in 1988, 1994, and 2004—revising the outlier payment regulations, which
are codified at 42 C.F.R. § 412.84. As presented below in greater detail, these regulations set a
formula for how individual outlier payments will be calculated for each fiscal year, a formula
that was revised over the course of time by these regulations. For all of the years after 1994,
including the years for which outlier payments are at issue in this litigation, that formula
involved a fixed loss threshold that was set annually. As described below in greater detail, the
fixed loss threshold represents the dollar amount of loss that a hospital must absorb in any case
in which the hospital incurs estimated costs for treating a patient above and beyond the payment
rate set for that type of case. Accordingly, the second set of regulations challenged in this action
consists of 11 annual fixed loss threshold rulemakings issued for FY 1997 to FY 2007. In each of
those annual rulemakings, the agency established a methodology for setting a fixed loss
4
threshold and set the dollar value of the fixed loss threshold itself. 4 The outlier payment
regulations that were applicable for the relevant fiscal year are effectively inputs for the fixed
loss threshold rulemakings. That is, using the methodology selected for the fiscal year, the
agency uses the applicable outlier payment regulations together with selected past data to
generate an annual fixed loss threshold that complies with the statutory requirements.
Because of the wide-ranging nature of Plaintiffs’ challenge—challenging 11 annual fixed
loss threshold regulations and outlier payment regulations issued over the course of three
decades—and because of the technical complexity of the program involved—it is necessary to
review the history of the program in some detail. 5 Moreover, it is important that significant
changes to the program occurred over the course of the years covered in this challenge. Because
Plaintiffs challenge rulemakings that occurred as early as 1988, it is necessary to explain the
changes that occurred as the outlier payment program unfolded over time. That said, the Court
provides the greatest detail on fiscal years 1997 through 2007 that are the core of the claims in
this case. The Court reserves certain details for its discussion of the discrete issues presented by
the parties below.
The Statutory Framework
Medicare “provides federally funded health insurance for the elderly and disabled,”
Methodist Hosp. of Sacramento v. Shalala, 38 F.3d 1225, 1226-27 (D.C. Cir. 1994), through a
4
For FY 2007, the agency only set a tentative final fixed loss threshold in the rule that finalized
the methodology for setting the fixed loss threshold because of uncertainty about certain other
data. See infra note 16. The agency later set the actual final fixed loss threshold for that year. See
id. That distinction is of no moment to this litigation.
5
The Court notes that, by contrast, Plaintiffs persist in taking a blunderbuss approach to this
challenge, despite the Court’s previous opinions indicating that doing so was not permissible,
eliding certain key differences between the variety of regulations they challenge and the years in
which they were promulgated.
5
“complex statutory and regulatory regime,” Good Samaritan Hosp. v. Shalala, 508 U.S. 402
(1993). The program is administered by the Secretary through the Centers for Medicare and
Medicaid Services (“CMS”). Cape Cod Hosp. v. Sebelius, 630 F.3d 203, 205 (D.C. Cir. 2011).
From its inception in 1965 until 1983, Medicare reimbursed hospitals based on “the
‘reasonable costs’ of the inpatient services that they furnished.” Cnty. of Los Angeles v. Shalala,
192 F.3d 1005, 1008 (D.C. Cir. 1999) (quoting 42 U.S.C. § 1395f(b)), cert. denied, 530 U.S.
1204 (2000). However, “[e]xperience proved … that this system bred ‘little incentive for
hospitals to keep costs down’ because ‘[t]he more they spent, the more they were reimbursed.’”
Id. (quoting Tucson Med. Ctr. v. Sullivan, 947 F.2d 971, 974 (D.C. Cir. 1991)).
In 1983, with the aim of “stem[ming] the program’s escalating costs and perceived
inefficiency, Congress fundamentally overhauled the Medicare reimbursement methodology.”
Cnty. of Los Angeles, 192 F.3d at 1008 (citing Social Security Amendments of 1983, Pub. L. No.
98–21, § 601, 97 Stat. 65, 149). Since then, the Prospective Payment System, as the overhauled
regime is known, has reimbursed qualifying hospitals at prospectively fixed rates. Id. By
enacting this overhaul, Congress sought to “reform the financial incentives hospitals face,
promoting efficiency in the provision of services by rewarding cost[-]effective hospital
practices.” H.R. Rep. No. 98–25, at 132 (1983), reprinted in 1983 U.S.C.C.A.N. 219, 351.
Under the Prospective Payment System, Hospitals are reimbursed “based on the average
rate of ‘operating costs [for] inpatient hospital services.’” Dist. Hosp. Partners, L.P. v. Burwell,
786 F.3d 46, 49 (D.C. Cir. 2015) (quoting Cnty. of Los Angeles, 192 F.3d at 1008). “Because
different illnesses entail varying costs of treatment, the Secretary uses diagnosis-related groups
(DRGs) to ‘modif[y]’ the average rate.” Id. (quoting Cape Cod Hosp., 630 F.3d at 205). “A DRG
is a group of related illnesses to which the Secretary assigns a weight representing ‘the
6
relationship between the cost of treating patients within that group and the average cost of
treating all Medicare patients.’” Id. (quoting Cape Cod Hosp., 630 F.3d at 205–06). “To calculate
a specific reimbursement, the Secretary ‘takes the [average] rate, adjusts it [to account for
regional labor costs], and then multiplies it by the weight assigned to the patient’s DRG.’” Id.
(quoting Cnty. of Los Angeles, 192 F.3d at 1009) (alteration in original).
“Congress recognized that health-care providers would inevitably care for some patients
whose hospitalization would be extraordinarily costly or lengthy” and devised a means to
“insulate hospitals from bearing a disproportionate share of these atypical costs.” Cnty. of Los
Angeles, 192 F.3d at 1009. Specifically, Congress authorized the Secretary to make supplemental
“outlier” payments to eligible providers. Id. While relatively simple in concept—hospitals
receive additional payments for extremely high cost treatments—implementing the outlier
payment concept entails a complex process, which has evolved substantially in the more than
three decades since outlier payments were introduced. Because Plaintiffs challenge aspects of the
implementation of the scheme that cover all three decades, it is necessary to review how the
program has evolved over time.
Pursuant to the 1983 legislation, the program provided for outlier payments for “day
outliers” and “cost outliers.” Cnty. of Los Angeles, 192 F.3d at 1009 (citing 42 U.S.C.
§ 1395ww(d)(5)(A)(i)-(ii) (Supp. IV 1986)). Day outliers are those patients whose “length of
stay exceeded the mean length of stay for that particular DRG by a fixed number of days or
standard deviations.” Cnty. of Los Angeles, 192 F.3d at 1009 (citing 42 U.S.C.
§ 1395ww(d)(5)(A)(i) (Supp. IV 1986)). In other words, day outlier payments cover patients
with extraordinarily lengthy hospital stays. Day outlier payments for extraordinary lengthy stays
were subsequently eliminated by Congress and are not at issue in this litigation. See 42 U.S.C.
7
§ 1395ww(d)(5)(A)(i) (day outlier payments only available for discharges occurring “during
fiscal years ending on or before September 30, 1997”). As specified by Congress in 1983, cost
outlier payments covered situations where “a hospital’s cost-adjusted charges surpassed either a
fixed multiple of the applicable DRG prospective-payment rate or such other fixed dollar amount
that the Secretary established.” 6 Cnty. of Los Angeles, 192 F.3d at 1009 (citing 42 U.S.C.
§ 1395ww(d)(5)(A)(ii) (Supp. IV 1986)). The 1983 statute further specified that the amount of
the outlier payments “shall be determined by the Secretary and shall approximate the marginal
cost of care beyond the cutoff point applicable” to day outliers or to cost outliers. 42 U.S.C.
§ 1395ww(d)(5)(A)(iii); see also Cnty. of Los Angeles, 192 F.3d at 1009. Finally, the statute
provides that “[t]he total amount of the additional [outlier] payments made under this
subparagraph for discharges in a fiscal year 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). In County of Los
Angeles, the D.C. Circuit Court of Appeals approved the Secretary’s interpretation of that final
provision—regarding the 5 to 6 percent range—under which the agency was required to set the
outlier threshold in advance such that the projected outlier payments would be between 5 and 6
percent of the total projected payments. See 192 F.3d at 1020. The Court of Appeals concluded
6
As explained further below, this provision of the statute was subsequently amended such that
outlier payments are made for “discharges in fiscal years beginning on or after October 1,
1994”—that is, the first day of FY 1995—where “charge, adjusted to cost, … exceed the sum of
the applicable DRG prospective payment rate plus any amounts payable under subparagraphs (B)
and (F) plus a fixed dollar amount determined by the Secretary.” 42 U.S.C.
§ 1395ww(d)(5)(A)(iii) (emphasis added). It is the annually-set “fixed dollar amount” and
factors related to the agency’s method of setting that amount that are the core of the dispute in
this case.
8
that the agency was not required to retroactively adjust the outlier thresholds such that the actual
outlier payments would be between 5 and 6 percent of the total payments. See id. at 1019-20.
Since the introduction of the Prospective Payment System, the applicable statutory
provision has provided for cost outlier payments only when “charges, adjusted to cost,” exceed a
certain cutoff. 42 U.S.C. § 1395ww(d)(5)(A)(ii). Because hospitals “markup” their costs in their
billing, a calculation is necessary in order to estimate the actual costs of a given treatment based
on the amount charged by a medical facility. See Dist. Hosp. Partners, 786 F.3d at 50. Until
1988, the agency used a standard cost-to-charge ratio for all facilities. See 53 Fed. Reg. 38,476,
38,502 (Sept. 30, 1988) (“We currently determine the cost of the discharge to be equal to 66
percent of the billed charges for covered services based on the average ratio of operating costs to
charges for Medicare discharges nationwide.”). However, in 1988 the agency decided to shift to
using hospital-specific cost-to-charge ratios, reasoning that “[t]he use of hospital-specific cost-to-
charge ratios should greatly enhance the accuracy with which outlier cases are identified and
outlier payments are computed, since there is wide variation among hospitals in these cost-to-
charge ratios.” Id. at 38,503. Accordingly, the agency codified that change by adding the
following provision to the Code of Federal Regulations specifying that that hospital-specific
cost-to-charge ratios would be used:
The cost-to-charge ratio used to adjust covered charges is computed annually by
the intermediary for each hospital based on the latest available settled cost report
for that hospital and charge data for the same time period as that covered by the
cost report. Statewide cost-to-charge ratios are used in those instances in which a
hospital’s cost-to-charge ratio falls outside reasonable parameters. HCFA sets
forth these parameters and the Statewide cost-to-charge ratios in each year’s
annual notice of prospective payment rates published under § 412.8(b).
53 Fed. Reg. at 38,529 (providing text of provision codified at 42 C.F.R. § 412.84(h)). In
explaining the choice to use the “latest available settled cost report,” the Secretary reasoned that
“the hospital-specific cost-to-charge ratios should be developed using the most current and
9
accurate data available.” 53 Fed. Reg. at 38,507. Furthermore, the Secretary reasoned that,
“[w]hile the latest filed cost report represents the most current data, we have found that Medicare
costs are generally overstated on the filed cost report and are subsequently reduced as a result of
audit.” Id. The agency considered the range of reasonable cost-to-charge ratios to be “3.0
standard deviations (plus or minus) from the mean of the log distribution of cost-to-charge ratios
for all hospitals.” Id. The agency explained that it “believe[d] that ratios falling outside this range
are unreasonable and are probably due to faulty data reporting or entry.” Id. at 38,507-08. In
other words, pursuant to the 1988 regulation, to determine whether a hospital’s cases qualified
for outlier payments, the agency would use the ratio between the latest available settled cost
report—that is, the most recent audited cost report—and the associated charges. But if the cost-
to-charge ratio was either extremely high or extremely low in comparison to the other hospitals,
the agency would use the statewide average cost-to-charge ratio in determining whether outlier
payments were warranted. These two aspects of the formula for calculating outlier payments—
the use of the latest available settled cost reports and the statewide average default—remained
applicable until they were modified by regulation in 2003. See 68 Fed. Reg. 34,494, 34,497-500
(June 9, 2003); 42 C.F.R. § 412.84(i)(3) (2003).
In 1993, Congress amended the statutory provisions establishing the outlier payment
framework—for FY 1995 and beyond—through Section 13501(c) of the Omnibus Budget
Reconciliation Act of 1993 (Public Law 103-66). 59 Fed. Reg. 45,330, 45,368 (Sept. 1, 1994).
Previously a “hospital [could] receive payment for a cost outlier if the adjusted costs for a
discharge exceed the greater of a fixed dollar amount or a fixed multiple of the DRG payment for
the case.” Id. Pursuant to the 1993 legislation, for discharges on or after October 1, 1994, a
hospital can request an outlier payment when the charges, adjusted to cost, “exceed the sum of
10
the applicable DRG prospective payment rate … plus a fixed dollar amount determined by the
Secretary.” 7 42 U.S.C. § 1395ww(d)(5)(A)(ii); see also 59 Fed. Reg. at 45,368. The “fixed dollar
amount” is known as the fixed loss threshold. “In effect, this threshold ‘acts like an insurance
deductible because the hospital is responsible for that portion of the treatment’s excessive cost’
above the applicable DRG rate.” Dist. Hosp. Partners, 786 F.3d at 50 (quoting Boca Raton Cmty.
Hosp., Inc. v. Tenet Health Care Corp., 582 F.3d 1227, 1229 (11th Cir. 2009)). The fixed loss
threshold is set by the Secretary each fiscal year. Id. The agency must set that fixed loss
threshold such that the projected outlier payments are between 5 and 6 percent of the total
projected DRG-based payments for the applicable fiscal year. See Cnty. of Los Angeles, 192 F.3d
at 1020.
To implement the amended statutory provisions, the agency amended its outlier payment
regulations—promulgating the second regulation challenged by Plaintiffs in this action. For all
discharges on or after October 1, 1994, cost outlier payments would be available when charges,
adjusted to cost, “exceed the DRG payment for the case plus a fixed dollar amount.” 59 Fed.
Reg. 45,330, 45398 (Sept. 1, 1994) (amending 42 C.F.R. § 412.80). The agency acknowledged
that the language of the statutory amendment contained some ambiguity as to whether the new
formula was required for future discharges or whether it provided an optional alternative to the
previous outlier payment formula. See id. at 45,370. But the agency concluded that adopting the
new fixed loss cost methodology exclusively was both substantively appropriate and consistent
with the intent of Congress in amending the relevant statutory provisions. See id.
7
The 1993 Act also eliminated the day outlier payment provision, discussed above, for all
discharges after September 30, 1997. 59 Fed. Reg. at 45,367. The Act also specified that the day
outlier payments must be a declining percentage of the total outlier payments for FY 1995, FY
1996, and FY 1997. See id.
11
In this same regulation, the agency set the fixed loss threshold for FY 1995 at $20,500.
See id. at 45,407. Pursuant to section 1395ww(d)(5)(A)(iii), there is a requirement that the outlier
payment “approximate the marginal cost of care beyond the cutoff point.” For FY 1995, the
agency set the marginal cost factor for cost outliers at 80 percent. 8 59 Fed. Reg. at 45,407.
Accordingly, cost outlier payments would be provided where a charge, adjusted to cost, was
greater than the applicable DRG rate plus $20,500. In such a circumstance, the amount of the
cost outlier payment would be 80 percent of the difference between the charges, adjusted to cost,
and the sum of the DRG rate and $20,500. For FY 1995, the agency set the fixed loss threshold
such that “estimated outlier payments equal 5.1 percent of estimated DRG payments.” Id. The
5.1 percent level satisfies the statutory requirement that the predicted outlier payments be
between 5 and 6 percent of the total DRG payments. See Cnty. of Los Angeles, 192 F.3d at 1020.
In order to set the fixed loss threshold such that outlier payments would meet this 5.1 percent
level, it was necessary to model the outlier and DRG payments for the upcoming fiscal year.
Whereas the agency had used a charge inflation methodology through FY 1993, the agency used
a cost inflation methodology for FY 1995 as it had done the previous fiscal year. 9 See 59 Fed.
Reg. at 45,407. In modeling the outlier payments for FY 1994, the agency concluded that a cost
inflation methodology would be more accurate than a charge inflation methodology. 58 Fed.
Reg. 46,270, 46,347 (Sept. 1, 1993). Specifically, the agency reasoned that, because charges
were rising faster than costs and because the cost-to-charge ratios used could be “as much as 2
years old” due to the use of the latest available settled cost reports, a charge inflation
8
The marginal cost factor is not at issue in this litigation.
9
Prior to FY 1994, the agency had used a charge inflation methodology, which involved inflating
a prior year’s charges by the level of charge inflation, and then adjusting those inflated charges to
costs using hospital-specific cost-to-charge ratios. 59 Fed. Reg. at 45,407.
12
methodology could be leading to “overestimating outlier payments in setting the thresholds.” Id.
The agency continued to use the cost inflation methodology in FY 1995 and through FY 2002.
See 68 Fed. Reg. 45,346, 45,476 (Aug. 1, 2003). Under the cost inflation methodology, to
generate a data set of projected costs for FY 1995, the agency first took the FY 1993 charge data
and adjusted the charges by the applicable cost-to-charge ratios. Id. Then the agency inflated
those costs to account for two years of cost inflation. Id. The agency then used this data set to
determine what fixed loss threshold would generate projected outlier payments that were 5.1
percent of the total projected DRG payments for FY 1995.
The final prong of the statutory scheme is that the payments based on the DRG
prospective payment rates—non-outlier payments—are reduced each year by a percentage equal
to the percentage established by the Secretary for outlier payments for that year. See 42 U.S.C.
§ 1395ww(d)(3)(B). This reduction offsets, approximately, the cost of the outlier payments for
each year. Accordingly, for those years where the Secretary set the fixed loss threshold such that
outlier payments were projected to be 5.1 percent of the total DRG-based payments, the
payments based on the DRG prospective payment rates were reduced by 5.1 percent. See 62 Fed.
Reg. 45,966, 46,011 (Aug. 29, 1997) (“Thus, for example, we set outlier thresholds so that the
outlier payments for operating costs are projected to equal 5.1 percent of total DRG operating
payments, and we adjust the operating standardized amounts correspondingly. We do not set
aside a pool of money to fund outlier cases.”). However, there is no guarantee that the actual
total outlier payments will equal the reduction in DRG payments. See Cnty. of Los Angeles, 192
F.3d at 1019-20.
13
Outlier Payments and Fixed Loss Thresholds: FY 1997 through FY 2003
Because Plaintiffs challenge each of the fixed loss thresholds set for FY 1997 through FY
2007, the Court reviews each of the rulemakings in which those thresholds were set. As
explained above, under the Secretary’s interpretation of the statute, which has been upheld by the
United States Court of Appeals for the District of Columbia Circuit, “she must establish the fixed
[loss] thresholds beyond which hospitals will qualify for outlier payments” at the start of each
fiscal year. Cnty. of Los Angeles, 192 F.3d at 1009. For each of these fiscal years, the agency
modeled the expected payments for that upcoming fiscal year and set the fixed loss threshold at a
rate such that the level of outlier payments was predicted to be 5.1 percent of the anticipated total
payments based on the DRG rates, which is within the 5 to 6 percent range of total DRG-based
payments set by the statute. See Cnty. of Los Angeles, 192 F.3d at 1009. As explained in further
detail below, the year 2003 was a critical pivot point in the outlier payment program. Based on
the discovery of abusive charging practices by certain hospitals, the agency set out to change the
rules for outlier payments to curb these abuses, ultimately modifying the regulations governing
outlier payments in a regulation promulgated in June 2003. See Dist. Hosp. Partners, 786 F.3d at
51-52. Accordingly, for the annual fixed loss threshold rulemakings between FY 1997 and FY
2003, the agency applied the outlier payment regulations promulgated in 1988 and in 1994, both
discussed above. For the annual fixed loss threshold rulemakings for FY 2004 through FY 2007,
the agency applied the outlier payment regulations as modified by the June 2003 regulation. The
Court first reviews, briefly, each of the fixed loss threshold rulemakings for FY 1997 through FY
2003. After reviewing the 2003 changes to the outlier payment regulations, the Court turns to the
four challenged fixed loss threshold rulemakings following those changes, for FY 2004 through
FY 2007.
14
Fixed Loss Threshold Rulemaking for FY 199710
In a proposed rule published in the Federal Register on May 31, 1996, the agency
proposed a fixed loss threshold of $11,050 and proposed to maintain the marginal cost factor for
cost outliers at 80 percent. 11 61 Fed. Reg. 27,444, 27,495 (May 31, 1996). The agency calculated
the outlier thresholds so that outlier payments were projected to equal 5.1 percent of the total
projected DRG payments. Id. In a final rule promulgated on August 30, 1996, the agency
concluded that a fixed loss threshold of $9,700 was appropriate in order to satisfy the 5.1 percent
projection. See 61 Fed. Reg. 46,166, 46,228 (Aug. 30, 1996). The agency maintained the 80
percent marginal cost factor as proposed. Id. The agency concluded that a $9,700 fixed loss
threshold was appropriate because of updated cost inflation data. Specifically, the agency noted
that “[t]he latest available Medicare cost reports indicate that hospital cost per case decreased
from FY 1993 to FY 1994 as well as from FY 1994 to FY 1995.” Id. The agency concluded that
the data “suggests a continued trend in cost deflation,” whereas the agency had assumed zero
cost inflation in the proposed rule. Id. (emphasis in original). Accordingly, in order to project the
costs and the associated Medicare payments for FY 1997, the agency used a negative annual
inflation factor of 1.906 percent. Id. Specifically, the agency deflated the cost data for 1995 by
1.906 percent, twice, to generate a set of projected cost data for FY 1997. Because of the
10
Pursuant to the 1993 legislation amending the outlier payment program, FY 1997 was the final
year that the outlier program included both day outliers and cost outliers. The Court does not
address certain complexities of the fixed loss threshold calculation for FY 1997 that result from
the relationship of the day outlier provisions and the cost outlier provisions because the parties
have not indicated that any issues result from the inclusion of day outliers in the calculations for
FY 1997.
11
The agency proposed a lower fixed loss threshold of $10,075 for those hospitals that had not
yet entered the Prospective Payment System for capital related costs. See 61 Fed. Reg. at 27,495.
The Court does not discuss the lower thresholds set for such hospitals further because it does not
affect the analysis below, nor do the parties emphasize this distinction.
15
predicted cost deflation, it was necessary to lower the fixed loss threshold, in comparison to the
proposed threshold, so that the outlier payments would be 5.1 percent of the total DRG-based
payments. See id. at 46,228-29.
Fixed Loss Threshold Rulemaking for FY 1998
In a proposed rule issued on June 2, 1997, the agency proposed a fixed loss threshold for
cost outliers of $7,600. See 62 Fed. Reg. 29,902, 29,946 (June 2, 1997). The agency proposed to
maintain the marginal cost factor for cost outliers of 80 percent. Id. Once again, the agency set
the threshold so that the projected outlier payments would be 5.1 percent of the projected total
DRG-based payments. See id. The proposed fixed loss threshold was premised on continued cost
deflation—which the agency derived by analyzing cost data from previous years—and the
agency used an annual cost deflation factor of 1.969 percent to project costs for FY 1998. See id.
In a final rule issued August 29, 1997, the agency selected a fixed loss threshold of $11,050, such
that outlier payments were projected to be 5.1 percent of the projected total DRG-based
payments. 12 62 Fed. Reg. at 46,040. The increase in comparison to the proposed threshold was
due, in part, to amendments to the outlier payment provisions of the statute, which required
adjusting the outlier payment model to account for other factors not relevant to the issues in this
action. 13 Cf. Dist. Hosp. Partners, 786 F.3d at 50, n.1 (noting that these factors were immaterial
12
The agency retained the proposed marginal cost factor of 80 percent. 62 Fed. Reg. at 46,040.
13
Between the issuance of the proposed rule and the issuance of the final rule on August 29,
1997, Congress amended the applicable statutory provision such that the cost outlier threshold
would be “based on ‘the sum of the applicable DRG prospective payment rate plus any amounts
payable under subparagraphs (B) [IME payments] and (F) [DSH payments] plus a fixed dollar
amount determined by the Secretary.’” 62 Fed. Reg. at 46,040 (quoting Balanced Budget, Pub.
Law 105–33, August 5, 1997, 111 Stat. 251). IME payments are for indirect medical education,
and DSH payments are for disproportionate share hospitals. As a result of this change, the
agency no longer adjusted hospital costs to exclude these payments. See id. The IME and DSH
payments themselves are not at issue in this action.
16
to the analysis of challenges to outlier payments). Based on more recent data available at the
time of the promulgation of the final rule, which showed continued declining costs, the agency
used an annual cost inflation factor of minus 2.005 percent. See 62 Fed. Reg. at 46,041.
Accordingly, for FY 1998, the fixed loss threshold was set at $11,050.
Fixed Loss Threshold Rulemaking for FY 1999
In a proposed rule issued on May 8, 1998, the agency proposed a fixed loss threshold for
cost outliers of $11,350. 63 Fed. Reg. 25,576 (May 8, 1998). The agency proposed maintaining
the marginal cost factor of 80 percent. Id. The agency calculated the proposed outlier threshold
so that the projected outlier payments would be 5.1 percent of the total projected DRG-based
payments. Id. As in previous years, these calculations were made using a cost inflation
methodology. See id. at 25,611. In the agency’s proposed rule, the projections were based on an
annual cost inflation factor of minus 1.831 percent. See id. In a final rule promulgated on July
31, 1998, the agency established a fixed loss threshold for FY 1999 of $11,100 and maintained
the marginal cost factor at 80 percent. 63 Fed. Reg. 40,954, 41,008 (July 31, 1998). The agency
continued to use a cost inflation methodology but used an updated cost inflation factor of minus
1.724 percent, which was suggested by the more recent data available at the time of the
promulgation of the final rule. See id. The agency calculated that the projected cost outlier
payments using the fixed loss threshold of $11,100 would be 5.1 percent of the total projected
DRG-based payments. See id.
Fixed Loss Threshold Rulemaking for FY 2000
In a proposed rule issued on May 7, 1999, the agency proposed a fixed loss threshold of
$14,575. 64 Fed. Reg. 24,716, 24,754 (May 7, 1999). The agency also proposed maintaining the
marginal cost factor of 80 percent. Id. The agency calculated the proposed outlier threshold so
17
that the projected outlier payments would be 5.1 percent of the total DRG-based payments. Id.
Once again, these calculations were made using a cost-inflation methodology. See id. For FY
2000, the agency proposed using a zero inflation factor. The use of a zero inflation factor
“reflects [the agency’s] analysis of the best available cost report data as well as calculations
(using the best available data) indicating that the percentage of actual outlier payments for FY
1998, is higher than [] projected before the beginning of FY 1998, and that the percentage of
actual outlier payments for FY 1999 will likely be higher than [] projected before the beginning
of FY 1999.” Id. In a final rule promulgated on July 30, 1999, the agency established a fixed loss
threshold for FY 2000 of $14,050 and maintained the marginal cost factor at 80 percent. 64 Fed.
Reg. 41,490, 41,546 (July 30, 1999). The agency used a zero inflation factor, as proposed,
reflecting the latest cost report data, as well as the relationship of actual outlier payments to the
previously projected outlier payments for FY 1998 and FY 1999, as discussed in the FY 2000
proposed rule. Id. The agency noted that it was “attempting to improve [its] estimate of payments
for FY 2000 by using a cost inflation factor of zero percent rather than a negative inflation
factor,” as the agency had done for several years prior to FY 2000. Id. at 41,547.
Fixed Loss Threshold Rulemaking for FY 2001
In a proposed rule issued on May 5, 2000, the agency proposed a fixed loss threshold of
$17,250. 65 Fed. Reg. 26,282, 26,329 (May 5, 2000). The agency also proposed maintaining the
marginal cost factor of 80 percent. Id. The agency calculated the proposed outlier threshold so
that the projected outlier payments would be 5.1 percent of the total DRG-based payments. Id.
Once again, these calculations were made using a cost-inflation methodology. See id. For FY
2001, the agency proposed using a 1.0 percent inflation factor. The use of a 1.0 percent inflation
factor “reflects [the agency’s] analysis of the best available cost report data as well as
18
calculations (using the best available data) indicating that the percentage of actual outlier
payments for FY 1999, is higher than [] projected before the beginning of FY 1999, and that the
percentage of actual outlier payments for FY 2000 will likely be higher than [] projected before
the beginning of FY 2000.” Id. In a final rule promulgated on August 1, 2000, the agency
established a fixed loss threshold for FY 2001 of $17,550 and maintained the marginal cost
factor at 80 percent. 65 Fed. Reg. 47,054, 47,113 (Aug. 1, 2000). The agency used a 1.8 percent
inflation factor—as opposed to the proposed inflation factor of 1.0 percent—reflecting the latest
cost report data, as well as the relationship of actual outlier payments to the previously projected
outlier payments for FY 1999 and FY 2000, as discussed in the FY 2001 proposed rule. Id.
Fixed Loss Threshold Rulemaking for FY 2002
In a proposed rule issued on May 4, 2001, the agency proposed a fixed loss threshold of
$21,000. 66 Fed. Reg. 22,646, 22,726-27 (May 4, 2001). The agency also proposed maintaining
the marginal cost factor of 80 percent. Id. at 22,727. The agency calculated the proposed outlier
threshold so that the projected outlier payments would be 5.1 percent of the total DRG-based
payments. Id. Once again, these calculations were made using a cost-inflation methodology. See
id. For FY 2002, the agency proposed using a 5.5 percent inflation factor. The use of the 5.5
percent inflation factor “reflects [the agency’s] analysis of the best available cost report data as
well as calculations (using the best available data) indicating that the percentage of actual outlier
payments for FY 2000, is higher than [] projected before the beginning of FY 2000, and that the
percentage of actual outlier payments for FY 2001 will likely be higher than [] projected before
the beginning of FY 2001.” Id. In a final rule promulgated on August 1, 2001, the agency
established a fixed loss threshold for FY 2002 of $21,025 and maintained the marginal cost
factor at 80 percent. 66 Fed. Reg. 39,828, 39,941 (Aug. 1, 2001). The agency used a 2.8 percent
19
inflation factor—as opposed to the proposed inflation factor of 5.5 percent—reflecting the latest
cost report data, as well as the relationship of actual outlier payments to the previously projected
outlier payments for FY 2000 and FY 2001, as discussed in the FY 2002 proposed rule.
Fixed Loss Threshold Rulemaking for FY 2003: Shift from Cost Inflation to Charge Inflation
Methodology
In a proposed rule issued on May 9, 2002, the agency proposed a fixed loss threshold of
$33,450. 67 Fed. Reg. 31,404, 31,510 (May 9, 2002). The agency also proposed maintaining the
marginal cost factor of 80 percent. Id. at 22,727. The agency calculated the proposed outlier
threshold so that the projected outlier payments would be 5.1 percent of the total DRG-based
payments. Id. While these calculations were made using a cost-inflation methodology, the
agency proposed a calculation that differed from those employed in previous years because of
the data that was available at the time of the FY 2003 proposed rule. See id. The agency noted
that, previously, “inflation factors have been calculated by measuring the percent change in costs
using the two most recently available cost report files.” Id. For example, for FY 2002, those were
the FY 1998 and FY 1999 cost reports. Id. However, when the agency was proposing the
threshold for FY 2003, the FY 2000 cost reports were not available because of processing delays.
See id. Therefore, the agency proposed projecting the cost inflation factor by constructing an
averaging model based on the changes to the data available from FY 1995 through FY 1999, the
most recent cost data available. See id. The agency emphasized that this proposal was based on
the unavailability of the FY 2000 data. See id. As a result of the model, the agency proposed
applying a 15 percent cost inflation factor to the FY 2001 data to generate the projected FY 2003
data. See id.
After receiving a significant numbers of comments regarding the methodology proposed
for accounting for inflation—which was proposed in light of the unavailability of the FY 2000
20
cost data—the agency declined to adopt the proposed cost inflation methodology in the final FY
2003 fixed loss threshold regulation. 67 Fed. Reg. 49,982, 50,124 (Aug. 1, 2002). Instead, in the
final rule promulgated on August 1, 2002, the agency adopted a charge inflation methodology.
See id. The agency reasoned that, because of substantial increases in the growth of charges,
which were increasing faster than costs, predicting future payments using a charge inflation
methodology would be more accurate than using a cost inflation methodology. See id.
Accordingly, in order to determine the fixed loss threshold for FY 2003, the agency calculated
the rate of charge inflation from FY 1999 to FY 2000 and from FY 2000 to FY 2001 and applied
this two-year rate of inflation—a total of 17.6398 percent—to the FY 2001 charge data to
generate a set of projected FY 2003 charge data. See id. The agency adjusted these projected
charges to cost for use in modeling outlier payments for FY 2003. See id. Using this charge
inflation methodology, as well as continuing to use a marginal cost factor of 80 percent, the
agency set the fixed loss threshold at $33,560 for FY 2003. See id. With this fixed loss threshold,
the agency predicted that outlier payments would be 5.1 percent of the total DRG-based
payments. See id.
Changes to the Outlier Payment Program in 2003
By 2003, it appeared that the outlier payment system was breaking down. See Dist. Hosp.
Partners, 786 F.3d at 51. In particular, concerns emerged that hospitals “could manipulate the
outlier regulations if their charges were ‘not sufficiently comparable in magnitude to their
costs.’” Id. (quoting 68 Fed. Reg. 10,420, 10,423 (Mar. 5, 2003)). In February 2003, the
Secretary transmitted a draft interim final rule discussing several possible changes to the outlier
21
payment system to the Office of Management and Budget (“OMB”). 14 See AR (2003
amendments) 4417.338. Based on analysis of trends in the cost and charge data, the draft also
would have revised the previously-established fixed loss threshold for FY 2003, lowering it to
$20,760 for discharges occurring between the would-be date of publication of the interim final
rule and the end of FY 2003. See AR (2003 amendments) 4417.376. Specifically, the draft
highlighted 123 hospitals that had rapidly increased their charges between FY 1999 and FY 2001
and whose outlier payments were a disproportionately high percentage of their total DRG
payments. See AR (2003 amendments) 4417.373. However, the Secretary later abandoned this
draft rule and never published it in the Federal Register. See Dist. Hosp. Partners, 786 F.3d at 54,
58.
Instead, on March 5, 2003, the agency published a notice of proposed rulemaking that
proposed several of the same changes to the outlier payment system suggested in the draft
interim final rule. See 68 Fed. Reg. at 10,420. In the notice of proposed rulemaking, the agency
described how a hospital could take advantage of “the time lag between the current charges on a
submitted bill and the cost-to-charge ratio taken from the most recent settled cost report.” Id. at
10,423. As the D.C. Circuit Court of Appeals described in District Hospital Partners, “[a]
hospital knows that its cost-to-charge ratio is based on data submitted in past cost reports. If it
dramatically increased charges between past cost reports and the patient costs for which
reimbursement is sought, its cost-to-charge ratio would ‘be too high’ and would ‘overestimate
the hospital’s costs.’” 786 F.3d at 51 (quoting 68 Fed. Reg. at 10,423). Reviewing actual outlier
payments from several previous years, the agency discovered “123 hospitals whose percentage
14
By executive order, agencies taking “significant regulatory action” are required to send OMB
“[t]he text of the draft regulatory action” before publishing that action in the Federal Register.
Exec. Order No. 12,866 § 6(a)(3)(B)(i), 58 Fed. Reg. 51,735 (Sept. 30, 1993).
22
of outlier payments relative to total DRG payments increased by at least 5 percentage points”
from FY 1999 to FY 2001. 68 Fed. Reg. at 10,423. Furthermore, for those 123 hospitals, “the
mean rate of increase in charges was 70 percent.” Id. at 10,424. Yet, cost-to-charge ratios for
these hospitals, which were based upon cost reports from prior periods, declined by only 2
percent.” Id. (emphasis added). While charges were increasing rapidly, the cost-to-charge ratios
did not reflect this change because they were based on earlier time periods. These hospitals, with
rapidly increasing charges, have become called turbo-chargers. Dist. Hosp. Partners, 786 F.3d at
51.
To address these concerns, the agency proposed three changes to the outlier payment
regulations. First, whereas previously the regulations—promulgated in 1988—required using the
most recent settled cost report when determining the cost-to-charge ratios for hospitals, the
agency proposed to use either the tentatively settled cost report or the final settled cost report,
whichever would be from the later cost reporting period. Id. at 10,423-24. Because it can take 12
to 24 months between the tentative settling of a cost report and the final settling of that cost
report, allowing the use of the tentatively settled cost reports would allow more up-to-date data
to be used in determining the outlier payments. See id. at 10,424. Second, the agency proposed
that it would no longer default to statewide average cost-to-charge ratios for hospitals with low
cost-to-charge ratios. See id. Previously the regulations—again, promulgated in 1988—specified
that, for hospitals with extremely high or extremely low cost-to-charge ratios, the statewide
average cost-to-charge ratio would be used in determining outlier payments. See id. The agency
proposed that the statewide average would still be used for hospitals that had not yet filed cost
reports and situations where the hospital-specific cost-to-charge ratios would exceed the upper
threshold, id., which had previously been set at three standard deviations above the mean of the
23
log distribution of cost-to-charge ratios for all hospitals. See 53 Fed. Reg. at 38,507. But for
hospitals that had low cost-to-charge ratios, they would “receive their actual cost-to-charge
ratios, no matter how low their ratios fall.” 68 Fed. Reg. at 10,424. Third, the agency proposed
reconciling outlier payments after settled cost reports were issued. Id. at 10,424-25. The agency
proposed adding a provision to the regulations such that outlier payments would “become subject
to adjustment when hospitals’ cost reports are settled.” Id. at 10,425. Beyond three major
changes, in contrast to the draft interim final rule, the agency did not propose a mid-year
correction for the FY 2003 fixed loss threshold because of uncertainty regarding charging
practices during FY 2003. See id. at 10,426-27.
On June 9, 2003, the agency promulgated a final rule, which included the three major
changes to the outlier payment regime proposed in the March 2003 notice of proposed
rulemaking. See 68 Fed. Reg. 34,494, 34,497-503 (June 9, 2003). First, for all discharges on or
after October 1, 2003, the cost-to-charge ratios would be based on more recent data—based on
tentatively settled cost reports rather than on final settled cost reports. See id. at 34,497-99,
34,515 (codified at 42 C.F.R. § 412.84(i)(1)-(2)). Second, for discharges on or after August 8,
2003, the agency would not default to statewide averages for hospitals with low cost-to-charge
ratios. See id. at 34,500 (codified at 42 C.F.R. §§ 412.84(h) and (i)(1)). Third, for discharges on
or after August 8, 2003, outlier payments would “become subject to adjustment when hospitals’
cost reports coinciding with the discharge are settled.” 15 Id. at 34,504 (codified at 42 C.F.R.
§ 412.84(i)(4)). Finally, consistent with the agency’s suggestion in the notice of proposed
15
For discharges between August 8, 2003, and October 1, 2003, the outlier payments were only
partially subject to reconciliation in light of the fact that the use of tentatively settled cost reports
was not yet in effect. See 42 C.F.R. § 412.525(a)(4); 68 Fed. Reg. at 34,515.
24
rulemaking, the agency concluded that adjusting the fixed loss threshold for the remainder of FY
2003 was not warranted. See id. at 34,506.
Fixed Loss Threshold Rulemakings for FY 2004 through 2007
Plaintiffs challenge four annual fixed loss threshold rules—for FY 2004 through FY
2007—issued after the agency promulgated the 2003 regulations that changed the methodology
for calculating outlier payments. As with the fixed loss thresholds for FY 1997 through FY 2003,
the agency was required to set the fixed loss threshold in order that the projected outlier
payments would be between 5 and 6 percent of the total projected DRG-based payments for the
applicable fiscal years. See Cnty. of Los Angeles, 192 F.3d at 1020. Each of the annual fixed loss
thresholds for FY 2004 through FY 2007 was set using the outlier payment methodology updated
in 2003. In setting each of these thresholds, the agency used a charge inflation methodology,
which had been reintroduced in 2003. For each of these years, the agency set the fixed loss
threshold so as to result in outlier payments being 5.1 percent of total DRG-based payments and,
for each of these years, the agency adopted a marginal cost factor of 80 percent. See 68 Fed. Reg.
45,346, 45,478 (Aug. 1, 2003); 69 Fed. Reg. 48,916, 49,278 (Aug. 11, 2004); 70 Fed. Reg.
47,278, 47,495 (Aug. 12, 2005); 71 Fed. Reg. 47,870, 48,151 (Aug. 18, 2006). However, there
are differences in the setting of each of these fixed loss thresholds, particularly regarding the data
sets that were used to calculate the threshold for each year. The Court, therefore, reviews how the
fixed loss threshold was set for each of these fiscal years.
Fixed Loss Threshold Rulemaking for FY 2004
In a proposed rule issued May 19, 2003, the agency proposed setting the fixed loss
threshold for FY 2004 at $50,645. 68 Fed. Reg. 27,154, 27,235 (May 19, 2003). The agency
proposed using a charge inflation rate of 12.8083 percent annually (27.3 percent over two years).
25
Id. The agency also proposed maintaining the marginal cost factor at 80 percent. Id. Notably this
proposed rule was published prior to the promulgation of the revised outlier payment
regulations—finalized on June 9, 2003—and therefore used the methodology that was in place
before the 2003 changes to the outlier payment regulations. See id. In issuing the proposed fixed
loss threshold rule, the agency noted that “any final rule subsequent to the March 5, 2003
proposed rule that implements changes to the outlier payment methodology is likely to affect
how we will calculate the final FY 2004 outlier threshold.” Id. The agency further noted that “the
final FY 2004 threshold is likely to be different from this proposed threshold, as a result of any
changes subsequent to the March 5, 2003 proposed rule.” Id.
The final fixed loss threshold rule for FY 2004 was issued on August 1, 2003—two
months after the changes to the outlier payment regulations were promulgated. 68 Fed. at 45,346.
The final fixed loss threshold rule took account of the revised outlier payment regulations, which
would govern outlier payments for discharges occurring in FY 2004 and beyond. See id.
Specifically, using the charge inflation methodology that the agency had re-introduced in the FY
2003 rulemaking, the agency used charge data from 2002 and inflated it by a two-year charge
inflation rate to project charges for FY 2004. Id. The agency used “the 2-year average annual rate
of change in charges per case,” from FY 2000 to FY 2001 and from FY 2001 to FY 2002, which
was 12.5978 percent annually (or 26.8 percent over two years). Id. To calculate projected costs
for FY 2004, the agency adjusted the projected charges by hospital-specific cost-to-charge ratios.
In doing so, the agency implemented changes introduced by the 2003 revisions to the outlier
payment regulations. First, the agency adjusted its methodology to use more recent data to
generate the cost-to-charge ratios because, in issuing the outlier payments for FY 2004, the latest
tentatively settled cost reports would be used rather than the latest settled cost reports. See id.
26
Therefore, whereas the agency had previously used cost-to-charge ratios from the Provider
Specific File to calculate the fixed loss threshold, the agency instead “matched charges-per-case
to costs-per-case from the most recent cost reporting year” and “then divided each hospital’s
costs by its charges to calculate the cost-to-charge ratio for each hospital.” Id. Second, whereas
previously the agency had used a statewide average cost-to-charge ratio for hospitals with low
cost-to-charge ratios, the revised outlier payment regulations specified that the hospital-specific
cost-to-charge ratio would be used “no matter how low their ratios actually fall.” Id. The agency
used the resultant hospital-specific cost-to-charge ratios to adjust the projected FY 2004 charges
to costs for that fiscal year. See id.
The final change introduced by the 2003 outlier payment regulations was that outlier
payments would become subject to reconciliation “at the time of cost report final settlement if a
hospital’s actual … cost-to-charge ratios are found to be substantially different from the cost-to-
charge ratios used during that time period to make outlier payments.” Id. The agency noted that it
was “difficult to project which hospitals will be subject to reconciliation of their outlier payments
using available data.” Id. The agency also noted that “resources necessary to undertake
reconciliation will ultimately influence the number of hospitals reconciled.” Id. For all of those
reasons, it was “difficult to predict the number of hospitals that [would] be reconciled.”
Nonetheless, based on previous data, the agency “identified approximately 50 hospitals [it]
believe[d] will be reconciled.” Id. For these hospitals, the agency attempted to integrate the
effects of reconciliation in the predictive model for FY 2004 by using a modified methodology
for projecting outlier payments for FY 2004. See id. at 45,476-77. Using this revised
methodology, in light of the changes to the outlier payment regulations in 2003, the agency
established a fixed loss outlier of $31,000. Id. at 45,477.
27
Fixed Loss Threshold Rulemaking for FY 2005
In a proposed rule issued May 18, 2004, the agency proposed a fixed loss threshold of
$35,085. 69 Fed. Reg. 28,196, 28,376 (May 18, 2004). To arrive at this proposed threshold, the
agency once again used a charge inflation methodology. The agency took FY 2003 charge data
and inflated it two years—based on the two-year average annual rate of change in charges from
FY 2001 to FY 2002 and FY 2002 to FY 2003, which was 14.5083 percent annually. See id.
Pursuant to the 2003 changes to the outlier payment regulations, the agency then used hospital-
specific cost-to-charge ratios that reflected the most recent settled or the most recent tentatively
settled cost reports, whichever was more recent, for each hospital. See id. In proposing this
threshold, the agency acknowledged that the inflation data “derive[d] from the years just prior to
the adoption of the policy changes, when some hospitals were increasing charges at a rapid rate
in order to increase their outlier payments.” Id. As a result, the agency noted that they “represent
rates of increase that may be higher than the rates of increase under our new policy.” Id.
Accordingly, the agency welcomed suggestions that might enable predictions that “better reflect
current trends in charge increases.” Id.
Given these concerns, as well as comments received in response to the proposed rule, the
agency adjusted the methodology for predicting FY 2005 data in order to set the final FY 2005
fixed loss threshold. See 69 Fed. Reg. at 49,277. In the final rule promulgated on August 11,
2004, the agency used the FY 2003 charge data as a baseline. Id. To derive an inflation factor, the
agency took the “unprecedented step of using the first half-year of data from FY 2003 and
comparing this data to the first half year of FY 2004.” Id. The agency concluded that “this
comparison will result in a more accurate determination of the rate of change in charges per case
between FY 2003 and FY 2005.” Id. Using this data, the agency calculated a one-year annual rate
of charge inflation of 8.9772, or 18.76 percent over two years, and then inflated the FY 2003 data
28
by this two-year charge inflation figure. Id. The agency then converted the projected charge data
to costs using hospital-specific cost-to-charge ratios from the April 2004 update of the Provider
Specific File. Id. As with the calculation of the cost-to-charge ratios for the setting of the FY
2004 fixed loss threshold, this analysis incorporated the changes introduced by the 2003
revisions to the outlier payment regulations, specifically the use of tentatively settled cost reports
and the use of hospital-specific ratios even for hospitals with extremely low cost-to-charge ratios.
See id. However, unlike the FY 2004 fixed loss threshold calculation, the agency did not change
its methodology to explicitly account for the possibility of reconciliation in setting the FY 2005
threshold. See id. at 49,278. The agency concluded that “due to changes in hospital charging
practices following implementation of the new outlier regulations in the June 9, 2003 final rule,
the majority of hospitals’ cost-to-charge ratios will not fluctuate significantly enough between
the tentatively settled cost report and the final settled cost report to meet the criteria to trigger
reconciliation of their outlier payments.” Id. Moreover, the agency noted that it would be
“difficult to predict which specific hospitals may be subject to reconciliation in any given year.”
Id. Using this methodology, the agency established a fixed loss threshold of $25,800 for FY
2005. Id.
Fixed Loss Threshold Rulemaking for FY 2006
In a proposed rule issued May 4, 2005, the agency proposed a fixed loss threshold of
$26,675. 70 Fed. Reg. 23,306, 23,469 (May 4, 2005). In order to account for changes in the rate
of charge inflation, the agency generated projected data for FY 2006 using FY 2004 charge data
as a baseline. See id. The agency calculated the rate of charge inflation from the combination of
the last quarter of FY 2003 and the first quarter of FY 2004 to the combination of the last quarter
of FY 2004 and the first quarter of FY 2005. See id. For this period, the agency calculated an
29
annual rate of charge inflation of 8.65 percent, or 18.04 percent over two years. See id. The
agency then inflated the FY 2004 charge data by this two-year inflation factor. See id. To derive
projected cost data, the agency adjusted the projected charge data by the hospital-specific charge
ratios from the most recent update to the Provider Specific File (for December 2004), which
reflected the 2003 changes to the outlier payment regulations. See id.
For the final rule, promulgated August 12, 2005, the agency calculated the FY 2006 fixed
loss threshold “using the methodology proposed in the proposed rule, but using updated data.” 70
Fed. Reg. at 47,494. Specifically, the agency calculated a charge inflation factor based on a
comparison between the first six months of FY 2004 and the first six months of FY 2005. Id.
Using this data, the agency calculated a two-year charge inflation factor of 14.94 percent. Id. The
agency used this inflation factor to inflate charges from FY 2004, id. at 47,495, and then adjusted
the projected charges to projected costs using hospital-specific cost-to-charge ratios from the
March 2005 update of the Provider Specific File, which included cost-to-charge ratios based on
the most recent tentatively settled cost reports, id. at 47,494. Just as in calculating the FY 2005
fixed loss threshold, the agency did not adjust its methodology to explicitly account for
reconciliation in establishing the FY 2006 threshold. Id. at 47,495. Based on this methodology,
the agency established a fixed loss threshold of $23,600. Id.
Fixed Loss Threshold Rulemaking for FY 2007
In a proposed rule issued on April 25, 2006, the agency proposed a fixed loss threshold of
$25,530. 71 Fed. Reg. 23,996, 24,150 (Apr. 25, 2006). The agency proposed to use the same
methodology to calculate the fixed loss threshold as it had in the past. Specifically, it proposed to
inflate FY 2005 charge data by two years of charge inflation. Id. at 24,149. The agency
calculated the rate of charge inflation by calculating the one-year average annual rate of change
30
from the combination of the last quarter of FY 2004 with the first quarter of FY 2005 to the
combination of the last quarter of FY 2005 with the first quarter of FY 2006. Id. at 24,149-50.
For this period, the agency calculated a one-year inflation rate of 7.57 percent, or 15.15 percent
over two years. Id. at 24,150. The agency inflated the FY 2005 charge data by this two-year rate
and then adjusted it to cost using the hospital-specific cost-to-charge ratios from the most recent
update to the Provider Specific File. Id.
For the final rule, promulgated on August 18, 2006, the agency calculated the FY 2007
fixed loss threshold “using the same methodology [] proposed, except [] using more recent data
to determine the charge inflation factor.” 71 Fed. Reg. at 48,150. In addition, however, the
agency also “appl[ied] an adjustment factor to the [cost-to-charge ratios] to account for cost and
charge inflation.” Id. Specifically, the agency calculated a charge inflation factor based on a
comparison between the first six months of FY 2005 and the first six months of FY 2006. Id.
Using this data, the agency calculated a two-year charge inflation factor of 16.42 percent. Id. The
agency used this inflation factor to inflate charges from FY 2005. Id. As in previous years, the
agency used hospital-specific cost-to-charge ratios to adjust the projected FY 2007 charges to
cost. Id. However, in response to comments submitted, the agency agreed “that it is appropriate
to apply an adjustment factor to the [cost-to-charge ratios] so that the [cost-to-charge ratios] we
are using in our simulation more closely reflect the [cost-to-charge ratios] that will be used in FY
2007.” Id. The agency calculated the ratio between the one-year change in costs and the one-year
change in charges, and derived an adjustment factor of 0.9973. Id. The agency applied this
adjustment factor to the cost-to-charge ratios contained in the Provider Specific File, which itself
was based on the most recent tentatively settled cost reports. Id. The agency then used these
adjusted cost-to-charge ratios to adjust the projected FY 2007 charges, generating projected FY
31
2007 costs. See id. Finally, just as in calculating the FY 2005 and FY 2006 fixed loss thresholds,
the agency did not adjust its methodology to explicitly account for reconciliation in establishing
the FY 2007 threshold. Id. at 48,149. Following this methodology, the agency adopted a tentative
fixed loss threshold of $24,475. Id. at 48,151. 16
* * *
In sum, for each fiscal year between 1997 and 2007, the Secretary determined the fixed
loss threshold, a fixed dollar amount that, when added to the DRG prospective payment rate,
serves as the cutoff point triggering eligibility for outlier payments. See 42 U.S.C.
§ 1395ww(d)(5)(A)(ii), (iv); 42 C.F.R. § 412.80(a)(2)-(3). When a hospital’s approximate costs
actually incurred in treating a patient exceed the sum of the DRG prospective payment rate
applicable to that patient and the fixed loss threshold, the hospital would be eligible for an outlier
payment in that case. See 42 U.S.C. § 1395ww(d)(5)(A)(ii)-(iii); 42 C.F.R. § 412.80(a)(2)-(3). In
this way, the fixed loss threshold represents the dollar amount of loss that a hospital must absorb
16
The agency adopted the actual final dollar amount in a subsequent rule. See 71 Fed. Reg. at
48,151. As the agency explained in full:
[T]he wage index tables, rates, and impacts will not be final in this final rule
because we are yet to determine occupational mix adjusted wage indices.
Therefore, we are only able to provide tentative standardized amounts, relative
weights, offsets, and budget neutrality factors in this final rule. Once we have the
final occupational mix data, we will recalculate these amounts to reflect the final
occupational mix adjusted wage indices. The same circumstances apply to the
outlier threshold. Without final wage index data, final standardized amounts, final
offsets and final budget neutrality factors, we are only able to provide a tentative
fixed loss outlier threshold in this final rule. Subsequent to this final rule, we will
publish a final fixed loss outlier threshold that will be effective for discharges on
and after October 1, 2006 for FY 2007. However, in this final rule, we are
adopting as final the methodology we will use to calculate the final outlier fixed-
loss cost threshold.
Id. at 48,149. Although the final fixed loss threshold was set at a later time, the
methodology, which is at issue in this case, was final as of the promulgation of the
August 2006 regulation.
32
in any case in which the hospital incurs estimated actual costs for treating a patient that are above
the DRG prospective payment rate. The amount of the outlier payment is “determined by the
Secretary” and must “approximate the marginal cost of care” beyond the fixed loss threshold. 42
U.S.C. § 1395ww(d)(5)(A)(iii). During the time period relevant to this action, the implementing
regulations generally provided for outlier payments equal to eighty percent of the difference
between the hospital’s estimated operating and capital costs and the fixed loss threshold. See 42
C.F.R. § 412.84(k). Accordingly, an increase in the fixed loss threshold reduces the number of
cases that will qualify for outlier payments, as well as the amounts of such payments for
qualifying cases. In light of this scheme, Plaintiffs challenge the rulemakings revising the outlier
payment regulations and the annual fixed loss threshold rulemakings that are applicable to the
payment determinations—for FY 1997 through FY 2007—that are challenged in this action.
B. Procedural Background
The procedural history is long and complex, and the Court limits its presentation of the
procedural history to the relevant facts for resolving the motions before the Court. Additional
procedural history can be found in the Court’s previous opinions in this case. See Banner Health
v. Sebelius (“Banner Health I”), 797 F. Supp. 2d 97 (D.D.C. 2011); Banner Health v. Sebelius
(“Banner Health II”), 905 F. Supp. 2d 174 (D.D.C. 2012); Banner Health v. Sebelius (“Banner
Health III”), 945 F. Supp. 2d 1 (D.D.C. 2013).
Plaintiffs in this case challenge outlier payment determinations for fiscal years 1997
through 2007. As required by statute, Plaintiffs filed various appeals with the Provider
Reimbursement Review Board (“PRRB”), each challenging the Secretary’s final outlier payment
determinations for the fiscal years in question, and Plaintiffs requested expedited judicial review
in those appeals. All but two of Plaintiffs’ requests for expedited judicial review were made in a
33
consolidated filing dated June 25, 2010. See PRRB Rs. at 27, 172, 428, 647, 980, 1304, 1771,
2410, 2831, 3125, 3556, 4091, 4594, 5037, 5383. Two additional requests were filed, on June 25,
2010, and on September 7, 2011, that were subsequently consolidated in this action. See PRRB
Rs. at 5633; PRRB R. (Case No. 11-0026) at 211. The June 25, 2010, request identified the legal
question at issue as follows:
Whether the Medicare Outlier Regulations, and the “fixed loss thresholds”
thereunder established by the Secretary of Health and Human Services (the
“Secretary”), and its Center for Medicare and Medicaid Services (“CMS”) and as
in effect for the Appealed Years, are substantively and/or procedurally invalid?
PRRB Rs. at 29. The legal requests presented in the two other separate requests for expedited
review did not differ in any substantive matter from the consolidated request. See PRRB Rs. at
5633; PRRB R. (Case No. 11-0026) at 211. Because Plaintiffs’ administrative appeals called into
question the underlying validity of regulations promulgated by the Secretary, the PRRB
determined that it was without authority to resolve the matters raised and authorized expedited
judicial review pursuant to 42 U.S.C. § 1395oo(f)(1). See PRRB Rs. at 1-3; PRRB R. (Case No.
11-0026) at 208-10.
Plaintiffs commenced this action on September 27, 2010, claiming that this Court has
jurisdiction under the Medicare Act, 42 U.S.C. § 1395oo(f)(1), and the Mandamus Act, 28
U.S.C. § 1361. See Compl., ECF No. 1. On December 23, 2010, Plaintiffs filed an Amended
Complaint as a matter of right, which remains the operative iteration of the Complaint in this
action (with a minor exception discussed below). See Am. Compl., ECF No. 16. As this Court
has previously observed, Plaintiffs’ Amended Complaint was “sprawling,” containing over two
hundred paragraphs, spanning fifty-nine pages, and including two lengthy exhibits. See Banner
Health III, 945 F. Supp. 2d at 9-10. On January 28, 2011, the Secretary filed a motion to dismiss
Plaintiffs’ Amended Complaint, which this Court granted in part and denied in part. See Banner
34
Health I, 797 F. Supp. 2d at 97. Specifically, the Court first concluded that Plaintiffs’ allegations
were sufficient, for the pleading stage, to establish their standing to challenge the several actions
at issue in this case. See id. at 109. The Court noted that this conclusion was not intended to
foreclose the Secretary from raising standing arguments upon filing motions for summary
judgment, after the record had been developed. See id. However, the Court dismissed—for
failure to state a plausible claim for relief—Plaintiffs’ claims seeking payments under the
Mandamus Act, 28 U.S.C. § 1361, as well as Plaintiffs’ claims under the Medicare Act to the
extent that such claims relied on vague allegations challenging the Secretary’s “implementation”
and “enforcement” of the outlier payment system that are “unconnected to any discrete agency
action.” See id. at 118. The Court otherwise denied the Secretary’s motion to dismiss. Looking
forward, the Court concluded that, in light of the extraordinary breadth of the allegations in the
Amended Complaint, proceeding immediately to the filing of the administrative record and the
subsequent briefing of motions for summary judgment would not be the most expeditious
manner of proceeding in the action. Rather, in order to gain further clarity as to the precise
contours of Plaintiffs’ claims, the Court ordered Plaintiffs to file a “notice of claims,”
identifying, in bullet-point format, each circumscribed, discrete agency action that Plaintiffs
intended to challenge. Id. at 117-18.
On July 27, 2011, Plaintiffs filed their Notice of Claims, which enumerated the claims
Plaintiffs were bringing in this action. Plaintiffs’ Notice of Claims likewise listed among the
challenged agency actions “the Secretary’s directions, starting in late 2002, to CMS’s fiscal
intermediaries to reopen hospital cost reports only for purposes of reconciling and recovering
outlier overpayments, but not for purposes of reconciling and recovering outlier underpayments,
as set forth in the Secretary’s issuance, through CMS, of Program Memorandum A–02–122
35
(December 3, 2002), Program Memorandum A–02–126 (December 20, 2002), Program
Memorandum A–03–058 (July 3, 2003)[, and] Transmittal 707 (Medicare Claims Processing
Manual, Chapter 3, § 20.1.2.5(A)).” However, on November 26, 2012, the Court granted the
Secretary’s motion to dismiss all claims premised on this agency action because, among other
reasons, Plaintiffs failed to rebut the Secretary’s well-reasoned jurisdictional arguments and in
fact expressly disclaimed any intent to bring a direct challenge to reopening determinations or
instructions as such. See Banner Health I, 797 F. Supp. 2d at 97; Banner Health II, 905 F. Supp.
2d at 174. The Court held it would not allow Plaintiffs to achieve supplementation of the
administrative record by injecting the action with ill-defined claims, but rather, whether the
administrative record should be supplemented to include the CMS documents was a question
more appropriately addressed in the context of the Court’s consideration of Plaintiffs’ motion to
compel. Id. In light of the Notice of Claims, in addition to the outlier payment determinations
specific to each of the hospital plaintiffs, the remaining claims in this action may be succinctly
summarized as challenging the promulgation and implementation of the following agency
actions: three rulemakings revising the outlier payment regulations, which were promulgated in
1988, 1994, and 2003; and eleven rulemakings establishing the annual fixed loss threshold for
federal fiscal years 1997 through 2007. See Banner Health III, 945 F. Supp. 2d at 13.
After Plaintiffs filed their Notice of Claims, the Court ordered the Secretary to file the
complete administrative record—which did not include any record relating to Plaintiffs’
dismissed claims regarding the four directives issued by CMS—by December 14, 2011. See
Scheduling and Procedures Order (Aug. 19, 2011), ECF No. 29. The Secretary filed the
administrative records for the fourteen challenged agency actions on November 8, 2011,
November 23, 2011, December 14, 2011, and December 28, 2011. See Banner Health III, 945 F.
36
Supp. 2d at 13 (citing ECF filings). 17 On January 6, 2012, the Secretary supplemented these
productions with additional data in electronic form after the Court entered a protective order
pertaining to such data. All together, the administrative records filed by the Secretary constitute
over 10,000 pages of documents, as well as tens of thousands of megabytes of electronic data
that have been produced to Plaintiffs.
On February 22, 2012, Plaintiffs filed a motion to compel, challenging the completeness
of the administrative record, Pls.’ Mot. to Compel Def. to File the Complete Administrative R.
and to Certify Same, ECF No. 52, which the Court dismissed without prejudice to renew, in light
of the Secretary’s represented intent to file supplements to the record. See Min. Order (Feb. 24,
2012). The Secretary subsequently filed two additional supplements. See Def.’s Notice of Filing
Supplements to Admin. R., ECF No. 57; Def.’s Notice of Filing Supplements to Admin. R., ECF
No. 58; see also Def.’s Notice of Filing Certified Copies of Previously Filed Supplements to
Admin. R., ECF No. 59.
On March 23, 2012, Plaintiffs filed a renewed motion to compel, requesting that the
Court order the Secretary to file the “complete administrative record,” by supplementing the
records she had previously filed with various documents, including certain data files identified
by Plaintiffs and all other documents that were before the agency in connection with its
rulemakings, and further requesting that the Court order the Secretary to certify to the Court and
Plaintiffs that the administrative record is complete. See Pls.’ Mem. of P. & A. in Supp. of
Renewed Mot. to Compel Def. to File the Complete Admin. R. and to Certify Same, ECF No. 60,
at 37. After the filing of several supplements by the parties, the Court granted in part and denied
17
The Secretary also filed the records of Plaintiffs’ proceedings before the PRRB. See Banner
Health III, 945 F. Supp. at 13 n.14.
37
in part the motion. See Banner Health III, 945 F. Supp. 2d at 39. The Court granted the motion
and ordered supplementation with respect to all or part of eight of the 36 discrete items subject to
Plaintiffs’ motion.18 See id. at 24-39. The Court denied Plaintiffs’ motion in all other respects.
See id.
After the Secretary filed the required supplementation of the Administrative record, the
Court granted the Secretary’s [99] Motion for Leave to File Motion to Dismiss for Lack of
Subject Matter Jurisdiction. However, instead of allowing Defendant to file the motion
separately, the Court ordered Defendant to file the motion simultaneously and, in the alternative
to, its motion for summary judgment. See Order and Docket Entry, dated August 13, 3013, ECF
No. 102. On July 7, 2014, the Court granted in part and denied in part Plaintiffs’ [108] Motion
for Leave to Further Amend and Supplement First Amended Complaint. Specifically, the Court
denied Plaintiffs’ request to include claims that the Secretary’s failure to disclose the draft 2003
interim final rule violated 5 U.S.C. § 553, but granted Plaintiffs’ request to amend their
Complaint to include factual allegations concerning the draft interim final rule. See Banner
Health v. Burwell, 55 F. Supp. 3d 1, 3 (D.D.C. 2014). Specifically, the Court allowed Plaintiffs to
18
Specifically, the Court ordered supplementation regarding the following items: the 2003 draft
interim final rule (item 1), Banner Health III, 945 F. Supp. at 27; the Impact Files relating to the
1988 and 2003 amendments to the outlier payment regulations, including the underlying
assumptions and associated data (items 14 and 15), id. at 33-34; Tables 8a and 8b for the fixed
loss threshold regulations for FY 1997 through FY 2007 (item 16), id. at 34; letters from the
Medicare Payment Advisory Commission and the American Hospital Association relating to the
FY 2004 fixed loss threshold determination (items 22 and 24), id. at 35; the April 22, 2002 Joint
Letter from CMS’ Center for Medicare Management, Office of Financial Management to the
fiscal intermediaries (item 26), id. at 37-38; and Program Memoranda A-02-122 and A-02-126
(item 33), id. Subsequently, the Court granted in part and denied in part the Secretary’s Motion
for Reconsideration, ECF No. 93. See Mem. Order, dated July 30, 2013, ECF No. 96. The Court
granted the motion with respect to the cost-to-charge data underlying the Impact Files and
vacated the previous order to supplement the record with the source data for those files. See id. at
19. In all other respects, the Court denied the Motion for Reconsideration and did not disturb its
previous decisions regarding the supplementation of the administrative record. See id. at 20.
38
amend their Complaint to include the allegations contained in sub-paragraphs 198.5(a)-(e) of
their Proposed Amendments. See id. at 14.
On September 9, 2014, Defendant filed her [126] Motion to Dismiss for Lack of Subject
Matter Jurisdiction and for Summary Judgment, and Plaintiffs filed their [127] Motion for
Summary Judgment. That same day, Plaintiffs also filed their [128] Motion (Related to Their
Motion For Summary Judgment) for Judicial Notice and/or for Extra-Record Consideration of
Documents and Other Related Relief. The parties briefed these motions and both, subsequently,
filed notices of supplemental authority. These motions are now ripe for resolution.
II. LEGAL STANDARD
Judicial review of Plaintiffs’ claims under the Medicare Act rests on 42 U.S.C.
§ 1395oo(f)(1), which incorporates the APA. See 42 U.S.C. § 1395oo(f)(1) (“Such action[s] …
shall be tried pursuant to the applicable provisions under chapter 7 of Title 5.”); see also
Abington Crest Nursing & Rehab. Ctr. v. Sebelius, 575 F.3d 717, 719 (D.C. Cir. 2009).
A. Supplementation of the Record
The Administrative Procedure Act directs the Court to “review the whole record or those
parts of it cited by a party.” 5 U.S.C. § 706. This requires the Court to review “the full
administrative record that was before the Secretary at the time he made his decision.” Citizens to
Preserve Overton Park v. Volpe, 401 U.S. 402, 420 (1971), abrogated on other grounds by
Califano v. Sanders, 430 U.S. 99 (1977). Courts in this Circuit have “interpreted the ‘whole
record’ to include all documents and materials that the agency directly or indirectly considered
… [and nothing] more nor less.” Pac. Shores Subdivision, Cal. Water Dist. v. U.S. Army Corps of
Eng’rs, 448 F. Supp. 2d 1, 4 (D.D.C. 2006) (citation omitted). “In other words, the administrative
record ‘should not include materials that were not considered by agency decisionmakers.’” Id.
39
(citation omitted). “[A]bsent clear evidence, an agency is entitled to a strong presumption of
regularity, that it properly designated the administrative record.” Id. at 5.
“Supplementation of the administrative record is the exception, not the rule.” Pac.
Shores, 448 F. Supp. 2d at 5 (quoting Motor & Equip. Mfrs. Ass’n, Inc. v. EPA, 627 F.2d 1095,
1105 (D.C. Cir. 1979)); Franks v. Salazar, 751 F. Supp. 2d 62, 67 (D.D.C. 2010) (“A court that
orders an administrative agency to supplement the record of its decision is a rare bird.”). This is
because “an agency is entitled to a strong presumption of regularity, that it properly designated
the administrative record.” Pac. Shores, 448 F. Supp. 2d at 5. “The rationale for this rule derives
from a commonsense understanding of the court’s functional role in the administrative state[:]
‘Were courts cavalierly to supplement the record, they would be tempted to second-guess agency
decisions in the belief that they were better informed than the administrators empowered by
Congress and appointed by the President.’ ” Amfac Resorts, L.L.C. v. Dep’t of Interior, 143 F.
Supp. 2d 7, 11 (D.D.C. 2001) (quoting San Luis Obispo Mothers for Peace v. Nuclear
Regulatory Comm’n, 751 F.2d 1287, 1325-26 (D.C. Cir. 1984)). However, an agency “may not
skew the record by excluding unfavorable information but must produce the full record that was
before the agency at the time the decision was made.” Blue Ocean Inst. v. Gutierrez, 503 F. Supp.
2d 366, 369 (D.D.C. 2007). The agency may not exclude information from the record simply
because it did not “rely” on the excluded information in its final decision. Maritel, Inc. v. Collins,
422 F. Supp. 2d 188, 196 (D.D.C. 2006). Rather, “a complete administrative record should
include all materials that might have influenced the agency’s decision[.]” Amfac Resorts, 143 F.
Supp. 2d at 12 (citations omitted). “[W]hile it is true that data and analysis compiled by
subordinates may be properly part of the administrative record despite not having actually passed
before the eyes of the Secretary,” to be included in the Administrative Record, “the data or
40
analysis must be sufficiently integral to the final analysis that was considered by the [agency],
and the [agency’s] reliance thereon sufficiently heavy, so as to suggest that the decisionmaker
constructively considered it.” Banner Health III, 945 F. Supp. 2d at 28.
B. Motion to Dismiss for Lack of Subject Matter Jurisdiction
Federal Rule of Civil Procedure 12(h)(3) provides that “[i]f the court determines at any
time that it lacks subject-matter jurisdiction, the court must dismiss the action.” Fed. R. Civ. P.
12(h)(3). In assessing its jurisdiction over the subject matter of the claims presented, a court
“must accept as true all of the factual allegations contained in the complaint” and draw all
reasonable inferences in favor of the plaintiff, Brown v. District of Columbia, 514 F.3d 1279,
1283 (D.C. Cir. 2008) (internal quotation marks omitted), but courts are “not required … to
accept inferences unsupported by the facts alleged or legal conclusions that are cast as factual
allegations.” Rann v. Chao, 154 F. Supp. 2d 61, 64 (D.D.C. 2001). Ultimately, the plaintiff bears
the burden of establishing the Court’s subject matter jurisdiction, Arpaio v. Obama, No. 14-5325,
2015 WL 4772774, at *5 (D.C. Cir. Aug. 14, 2015), and where subject-matter jurisdiction does
not exist, “the court cannot proceed at all in any cause.” Steel Co. v. Citizens for a Better Env’t,
523 U.S. 83, 94 (1998).
C. Motions for Summary Judgment
Under Rule 56(a) of the Federal Rules of Civil Procedure, “[t]he court shall grant
summary judgment if the movant shows that there is no genuine dispute as to any material fact
and the movant is entitled to judgment as a matter of law.” However, “when a party seeks review
of agency action under the APA [before a district court], the district judge sits as an appellate
tribunal. The ‘entire case’ on review is a question of law.” Am. Bioscience, Inc. v. Thompson, 269
F.3d 1077, 1083 (D.C. Cir. 2001). Accordingly, “the standard set forth in Rule 56[ ] does not
41
apply because of the limited role of a court in reviewing the administrative record.... Summary
judgment is [ ] the mechanism for deciding whether as a matter of law the agency action is
supported by the administrative record and is otherwise consistent with the APA standard of
review.” Southeast Conference v. Vilsack, 684 F. Supp. 2d 135, 142 (D.D.C. 2010).
The APA “sets forth the full extent of judicial authority to review executive agency
action for procedural correctness.” FCC v. Fox Television Stations, Inc., 556 U.S. 502, 513
(2009). It requires courts to “hold unlawful and set aside agency action, findings, and
conclusions” that are “arbitrary, capricious, an abuse of discretion, or otherwise not in
accordance with law.” 5 U.S.C. § 706(2)(A). “This is a ‘narrow’ standard of review as courts
defer to the agency’s expertise.” Ctr. for Food Safety v. Salazar, 898 F. Supp. 2d 130, 138
(D.D.C. 2012) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co.,
463 U.S. 29, 43 (1983)). An agency is required to “examine the relevant data and articulate a
satisfactory explanation for its action including a rational connection between the facts found and
the choice made.” Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43 (internal quotation omitted).
“Moreover, an agency cannot ‘fail[ ] to consider an important aspect of the problem’ or ‘offer[ ]
an explanation for its decision that runs counter to the evidence’ before it.” Dist. Hosp. Partners,
786 F.3d at 57 (quoting Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43). The reviewing court “is not
to substitute its judgment for that of the agency.” Id. Nevertheless, a decision that is not fully
explained may be upheld “if the agency’s path may reasonably be discerned.” Bowman Transp.,
Inc. v. Arkansas–Best Freight Sys., Inc., 419 U.S. 281, 286 (1974).
III. DISCUSSION
This action entails Plaintiffs’ challenges to 14 rulemakings, three rulemakings revising
the outlier payment regulations and 11 annual fixed loss threshold rulemakings. The parties filed
42
cross-motions for summary judgment, and Defendant moved to dismiss for lack of subject matter
jurisdiction with respect to certain claims. In addition, Plaintiffs moved the Court to take judicial
notice and/or extra-record consideration of certain other materials, as well as specified other
relief. The Court begins by considering Plaintiffs’ motion requesting the consideration of
additional materials because the resolution of that motion determines the scope of the materials
examined with respect to the other motions resolved today. The Court then turns to Defendant’s
arguments that this Court lacks subject matter jurisdiction over certain claims. Finally, having
resolved those issues, the Court addresses the cross-motions for summary judgment and resolves
the merits of this dispute.
A. Plaintiffs’ Motion Regarding Consideration of Additional Materials
On the same date that the parties filed their dispositive cross-motions—on the deadline
previously set by the Court for those cross-motions, see Scheduling and Procedures Order, ECF
No. 122—Plaintiffs filed a motion asking the Court to consider additional materials in resolving
the cross-motions filed that day. Plaintiffs asked the Court (1) to take judicial notice of six
identified documents, or in the alternative to consider them as extra-record materials, (2) to
supplement the administrative record with a comment to the FY 2004 fixed loss threshold
rulemaking, and (3) to “receive” three tables as demonstrative exhibits that exceed the allowed
page limits set by the Court for the briefing. Defendant opposes each of these requests.
As an initial matter, the Court notes that it considers this request to be tardy. Asking the
Court for permission to consider additional materials on the very day on which the dispositive
motions are filed is simply too late. Doing so meant that Plaintiffs precluded Defendant from
effectively objecting to the inclusion of these materials before Plaintiffs relied on them in their
briefing. At the same time, by waiting until the last moment to present these materials, Defendant
43
was prevented from being able to rely on them in her initial briefing. To say that Defendant could
file an opposition to this motion after the fact and could respond to the proposed materials in the
subsequent briefing is no answer at all. The orderly management of this litigation requires the
resolution of questions regarding the materials on which the parties may rely—and on which the
Court will base its ultimate decision—before the submission of the parties’ arguments regarding
the merits. Last minute surprises regarding materials on which Plaintiffs hope to rely are neither
efficient nor fair. With that in mind, the Court considers each of Plaintiffs’ requests.
Judicial Notice or Extra-record Consideration of Six Documents
Plaintiffs ask the Court to take judicial notice “as a matter of adjudicative fact” or, in the
alternative, to consider as extra-record evidence six documents (falling into four categories): (1)
two versions of a statement given by Thomas A. Scully, former Administrator of the Centers for
Medicare and Medicaid Services (“CMS”), on March 11, 2003, before a Senate subcommittee;
(2) a report of the Office of Inspector General of the Department of Health and Human Services
regarding the reconciliation of outlier payments 19; (3) the declaration of Peter Reisman, a CMS
official, that was submitted in a case filed in the Southern District of New York; and (4) two
amicus briefs submitted by the United States in a case relating to outlier payments in the
Southern District of Florida.
The Court agrees—as does Defendant—that parties may cite to publicly available
documents and that the Court may take judicial notice of such documents. See Pharm. Research
& Manufacturers of Am. v. United States Dep’t of Health & Human Servs., 43 F. Supp. 3d 28, 33
19
The full title of the report in question is as follows: The Centers for Medicare and Medicaid
Services Did Not Reconcile Medicare Outlier Payments In Accordance with Federal Regulations
and Guidance (2012), available at https://oig.hhs.gov/oas/reports/region7/71002764.pdf (last
visited July 23, 2015).
44
(D.D.C. 2014). Each of the identified documents is publicly available, and the Court considers it
proper to take judicial notice of them. While Plaintiffs frame their request for judicial notice as
an alternative to their request for extra-record consideration of these documents, Plaintiffs have
revealed their hand. It appears that, regardless of Plaintiffs’ framing, Plaintiffs are effectively
asking the Court to consider these materials as extra-record evidence serving as the basis for the
Court’s review of the agency actions challenged here. Insofar as Plaintiffs seek to base their
challenge upon these extra-record materials, even those available to the public of which the
Court could take judicial notice, the Court concludes that it is necessary to apply the standard for
considering extra-record evidence.
It is “black-letter administrative law that in an [Administrative Procedure Act] 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)
(quoting Walter O. Boswell Mem’l Hosp. v. Heckler, 749 F.2d 788, 792 (D.C. Cir. 1984)). As the
D.C. Circuit Court of Appeals has recently reiterated, “exceptions to that rule are quite narrow
and rarely invoked.” CTS Corp. v. EPA, 759 F.3d 52, 64 (D.C. Cir. 2014). “They are primarily
limited to cases where ‘the procedural validity of the agency’s action remains in serious
question,’ or the agency affirmatively excluded relevant evidence.” Id. (citations omitted).
Plaintiffs have not even attempted to show that this standard is met in this case. Plaintiffs
emphasize that these documents should be considered because they are “needed to assist the
court’s review.” Pls.’ Extra-Record Mot. at 9 (quoting Nat’l Mining Ass’n v. Jackson, 856 F.
Supp. 2d 150, 156-57 (D.D.C. 2012). But that is not enough. As the D.C. Circuit has emphasized,
consideration of such extra-record materials “at most [] may be invoked to challenge gross
procedural deficiencies—such as where the administrative record itself is so deficient as to
45
preclude effective review.” Hill, 709 F.3d at 47. This is not such a case. Plaintiffs are not using
these materials to “challenge gross procedural deficiencies.” CTS Corp., 759 F.3d at 64. While
some procedural deficiencies undoubtedly exist in this case—such as missing comments from
the FY 2004 fixed loss threshold rulemaking—the Court has adequately accounted for those
deficiencies in resolving the parties’ previous motions regarding the scope of the administrative
record. See Banner Health III, 945 F. Supp. 2d at 19. Contrary to Plaintiffs’ characterization, the
record readily permits “effective review” without adding the requested materials to the record.
The Court concludes that, pursuant to the standard for consideration of extra-record evidence, it
would not be proper to consider the presented materials along with the already-voluminous
administrative record. In other words, Plaintiffs cannot evade that strict standard by appealing to
the standard for judicial notice. Although the Court will take judicial notice of these documents
as necessary in resolving this matter, the documents cannot serve as the foundation for Plaintiffs’
claims.
Supplementation with Federation of American Hospitals Comment
Plaintiffs seek to supplement the administrative record for the FY 2004 fiscal loss
threshold rulemaking with a comment by the Federation of American Hospitals. The Court
concludes that it is too late to—once again—expand the administrative record in this case. The
Court previously set a deadline of March 23, 2012, for Plaintiffs’ motion to compel regarding the
contents of the administrative record in this case. See Minute Order dated February 24, 2012.
After numerous supplemental filings, the Court resolved Plaintiffs’ motion to compel on May 16,
2013, addressing a wide variety of challenges to the administrative record filed. See Banner
Health, 945 F. Supp. 2d at 6. Subsequently, the Court granted Defendant’s [93] Motion for
Partial Reconsideration or Clarification and modified its previous order compelling the
46
supplementation of the administrative record with respect to certain materials. See Order dated
July 30, 2013, ECF No. 96. Meanwhile, on September 19, 2013, another judge in this district
accepted the comment in question into the administrative record in an outlier case proceeding in
parallel to this one. See Dist. Hosp. Partners, L.P. v. Sebelius, 971 F. Supp. 2d 15, 27 (D.D.C.
2013) aff’d in part and rev’d in part sub nom. Dist. Hosp. Partners, L.P. v. Burwell, 786 F.3d 46
(D.C. Cir. 2015). It is undisputed that the parties in this case were aware of the proceedings in
District Hospital Partners; indeed, from time to time, the parties reported to this Court regarding
those parallel proceedings, particularly regarding disputes over the administrative record in that
case. See, e.g., Plaintiffs’ Notice of the Conclusion of Previously Reported Discovery
Proceedings in the District Hospital Partners Case, dated April 12, 2013, ECF No. 81. Despite
the fact that Federation of American Hospitals comment was added to the record in District
Hospital Partners approximately one year before the dispositive motions in this case were due,
Plaintiffs never sought leave to add the Federation of American Hospitals comment to this case
until the day they filed their dispositive motion.
Plaintiffs argue that Defendant cannot be prejudiced because the agency knew about this
comment, but that misses the point. Defendant had no reason to believe that Plaintiffs would—at
the very last minute—seek to add this particular comment to the record given that the
administrative record had been extensively litigated in this case and given the lag between the
addition of the comment in question to the District Hospital Partners record and the filing of the
dispositive motions in this case. Accordingly, the Court concludes that it would be improper to
add the Federation of American Hospitals comment to the administrative record. 20
20
In any event, as the Court explains below, even if the Federation of American Hospitals
comment were added to the administrative record, the outcome in this case would not be altered.
47
Three Additional Tables
Plaintiffs ask the court to “receive” three tables that are appended to Plaintiffs’ Motion for
Summary Judgment. Plaintiffs explain that “[a]ll three tables summarize data in the Impact Files
that are already part of the administrative record and concisely encapsulate what would
otherwise require resorting to a voluminous portion of the administrative records.” Pls.’ Extra-
Record Mot. at 7. They argue that, because these tables “merely restate information already
contained in the administrative record,” they should not be included in the page limitations
previously ordered by the Court. Id. at 7-8. Plaintiffs further state that the “tables are intended to
aid the parties and the Court by presenting visually concise data that would otherwise only be
accessible in cumbersome spreadsheets.” Id. at 8. Defendant responds that Plaintiffs’ request to
submit the tables was untimely, that the additional pages should not be exempt from the page
limit because they effectively present argument rather than materials from the administrative
record, and that the Court should not allow an expansion of the page limits. The Court agrees
with Defendant.
First, pursuant to Local Civil Rule 7(e), Plaintiffs should have requested leave to exceed
the page limits—or to submit what they believe should not be counted towards the page limits—
prior to the deadline for filing their briefs. Presenting this request simultaneous with the filing of
those briefs is neither fair—Defendant did not have an opportunity to file an expanded brief—
nor conducive to the orderly administration of this action.
Second, the Court concludes that the appended tables are not exempt from the page limit.
Contrary to Plaintiffs’ suggestion, the exhibits are not like a Joint Appendix. A Joint Appendix
contains excerpts of the unadorned administrative record to facilitate the Court’s review of those
portions of the administrative record on which the parties rely, which is particularly useful in
cases like this one with a voluminous record. See LCvR 7(n). A Joint Appendix does not provide
48
an opportunity for the parties to add analysis or commentary regarding the meaning of the
administrative record. The Court agrees that, as a general matter, charts or graphs may be a
helpful way to present complex information. See generally Edward Tufte, Beautiful Evidence
(2006). They also can be an effective way to present an argument. See id. However, the question
is not whether or not the charts are helpful. Any material describing the administrative record
that is not a faithful reproduction of the unadorned administrative record belongs in a
memorandum of points and authorities in support of, or in opposition to, a motion. Indeed, that is
essentially what a factual or background section of a memorandum is: it is a description of a
voluminous record to set the groundwork for a party’s legal arguments about the meaning of that
record. Whether the charts themselves are more like a background section of a brief or the
argument section itself is immaterial for present purposes; it is plain that they are not like
portions of the administrative itself and that they should be counted towards the page limit for
the briefing.
In that light, the Court arrives at the page limitation itself. The Court’s Scheduling and
Procedures Order, dated July 17, 2014, established a 70-page limit for Plaintiffs’ motion for
summary judgment. See ECF No. 122, at 3. In that Order, the Court also stated that the “page
limits identified above are similarly generous and equally firm,” just as the briefing schedule set
that day was generous and firm. Id. at 4. The Court also “caution[ed] the parties that any attempt
to subvert these page limits by including additional briefing in appendices will be rejected, and
such appendices will be stricken from the record.” Id. (citing Hajjar-Nejad v. George Washington
Univ., 37 F. Supp. 3d 90, 114 (D.D.C. 2014)). Despite their arguments to the contrary, Plaintiffs
have done precisely that. Plaintiffs submitted a memorandum of points and authorities in support
of their motion for summary judgment of 70 pages. In addition, they submitted three tables—
49
exhibits 5, 7, and 8—that purport to summarize relevant data in the administrative record. These
exhibits are properly considered portions of Plaintiffs’ memorandum and, therefore, exceed the
page limit.
Plaintiffs’ suggestion—as a basis for submitting the separate exhibits—that they could
not have effectively cited to the administrative record is immaterial. Insofar as they believe they
effectively cited to the record in their charts, and then relied on these charts in their briefing, they
could have simply included these charts within their briefing itself. 21 Parties always face
tradeoffs in choosing what to include in a page-limited filing. If Plaintiffs had concluded that
including the tables was valuable in presenting their arguments—more valuable than other
materials they did include—the Court is confident that Plaintiffs could have devised a way to
include those materials in their memorandum itself. Instead, Plaintiffs are trying to have it both
ways, but the Court will not allow them to do so. Accordingly, the Court will not grant Plaintiffs’
belated request to submit the tables, which is effectively a request to expand the page limit for
the briefing of this action. The Court denies Plaintiffs’ request to submit these additional exhibits
and strikes exhibits 5, 7, and 8 from the record.
Having resolved these preliminary issues, the Court proceeds to consider the Secretary’s
arguments that this Court does not have subject matter jurisdiction over certain claims in this
action.
21
It is similarly immaterial that they had space within the final page of the brief to list all of the
citations to the administrative record on which those three tables rely.
50
B. Secretary’s Motion to Dismiss for Lack of Subject Matter Jurisdiction
The Secretary argues that several of Plaintiffs’ claims should be dismissed for lack of
subject matter jurisdiction. The Court reviews each of those arguments in turn and concludes that
it has jurisdiction over the claims in this action.
First, the Secretary argues that Plaintiffs lack standing to pursue claims that rely on
speculation about how hospitals—other than Plaintiffs—would behave under a different set of
regulations. To satisfy the standing requirements of Article III, a plaintiff “must show (1) it has
suffered an ‘injury in fact’ that is (a) concrete and particularized and (b) actual or imminent, not
conjectural or hypothetical; (2) the injury is fairly traceable to the challenged action of the
defendant; and (3) it is likely, as opposed to merely speculative, that the injury will be redressed
by a favorable decision.” Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S.
167, 180-81 (2000). Essentially the Secretary argues that a subset of Plaintiffs’ claims—although
the Secretary does not identify specifically which claims—rely on assumptions about how
hospitals would, generally, respond to the regulations in question. Therefore, the Secretary
argues, the injury was not “fairly traceable” to the challenged agency action. Plaintiffs respond
that their injuries are directly linked to each of the challenged regulations. The Court agrees that
Plaintiffs have standing to challenge the 14 regulations at issue in this case.
As described above, in Banner Health I, in resolving the Secretary’s motion to dismiss,
the Court concluded that Plaintiffs’ allegations were sufficient, for the pleading stage, to establish
their standing over all of the claims that survived the motion to dismiss stage. 22 See 797 F. Supp.
2d at 109. At that time, the Court noted that this conclusion was not intended to foreclose the
Secretary from raising standing arguments upon filing a motion for summary judgment, after the
22
Whether Plaintiffs have standing over claims that the Court dismissed for failure to state a
claim is immaterial at this stage of the proceedings. See Banner Health I, 797 F. Supp. 2d at 109.
51
record had been developed. See id. The Secretary now does so. However, neither the subsequent
development of the record nor the gradual refining of Plaintiffs’ claims alters the Court’s
previous conclusion that Plaintiffs have standing. As the Court noted in resolving the Secretary’s
motion to dismiss, “[i]n many cases, a plaintiff's ‘standing to seek review of administrative
action is self-evident,’ including where, as here, ‘the complainant is an object of the action (or
forgone action) at issue—as is the case usually in review of a rulemaking and nearly always in
review of an adjudication.’” Id. at 108 (quoting Sierra Club v. Envtl. Prot. Agency, 292 F.3d 895,
900 (D.C. Cir. 2002)) (internal quotation marks and citation omitted). As the Court stated
previously, Plaintiffs certainly have standing to challenge the outlier payments that Plaintiffs
themselves received for FY 1997 through FY 2007 and, accordingly, Plaintiffs also have standing
to challenge the validity of the regulations on which those payments were based—i.e., three
rulemakings revising the outlier payment regulations and 11 annual fixed loss threshold
regulations. See id.
Essentially, the Secretary’s standing argument amounts to a claim that the challenged
regulations were not arbitrary or capricious, or otherwise not in accordance with law, because
Plaintiffs’ arguments rely on predictions about actions of third parties. As the Court said
previously, this argument is best considered on the merits because, even if the Secretary’s
arguments are correct, they would not undermine Plaintiffs’ standing to bring the challenges in
this action in the first instance. See id. at 108 n.11. Notwithstanding the subsequent development
of the record and of the parties’ arguments, the Court’s conclusion that the Plaintiffs have
standing to bring the claims in this action is unchanged.
Second, the Secretary argues that this Court has no subject matter jurisdiction over
Plaintiffs’ claim that it was arbitrary and capricious for the agency to fail to make a mid-year
52
adjustment to the FY 2003 fixed loss threshold because the PRRB had not approved that question
for judicial review. Plaintiffs respond that the question of the propriety of a mid-year adjustment
to the FY 2003 fixed loss threshold was within the scope of judicial review approved by
Plaintiffs. Regarding this question, the Court agrees with Plaintiffs.
Under the Medicare Act, “[j]udicial review may be had only after the claim has been
presented to the Secretary and administrative remedies have been exhausted.” Am. Chiropractic
Ass'n, Inc. v. Leavitt, 431 F.3d 812, 816 (D.C. Cir. 2005). In this case, Plaintiffs presented
requests for judicial review with respect to all of the payment determinations at issue in this case.
The question for which Plaintiffs requested judicial review was framed broadly:
Whether the Medicare Outlier Regulations, and the “fixed loss thresholds”
thereunder established by the Secretary of Health and Human Services (the
“Secretary”), and its Center for Medicare and Medicaid Services (“CMS”) and as
in effect for the Appealed Years, are substantively and/or procedurally invalid?"
PRRB Rs. 28. Plaintiffs’ request made it clear that they were challenging the 2003 rulemaking
promulgating amendments to the outlier payment regulations. See id. It is that rulemaking that
encompasses Plaintiffs’ mid-year adjustment claim—whereby Plaintiffs argue that the decision
not to lower the FY 2003 fixed loss threshold in a mid-year adjustment was arbitrary and
capricious. The Secretary argues that Plaintiffs’ request for judicial review does not include the
mid-year adjustment claim because Plaintiffs’ request included discussion of several other
specific questions—not including the mid-year adjustment question—and references to several
specific Federal Register pages—which did not include the pages on which the decision not to
adjust the FY 2003 fixed loss threshold was explained. However, the Court does not read
Plaintiffs’ request so narrowly so as to exclude a challenge to the decision to refrain from a mid-
year adjustment to the FY 2003 threshold.
53
The decision in District Hospital Partners, 794 F. Supp. 2d at 162, is not to the contrary.
In District Hospital Partners—another outlier case—another judge in this district concluded that
the court did not have subject matter jurisdiction over the claims of the hospital-plaintiffs in that
case, arguing that the agency had failed to adjust the fixed loss thresholds for FY 2004 through
FY 2006, because those plaintiffs had not presented the mid-year adjustment claims to the
agency, as required. See id. at 168-69. At the outset, it is important that different mid-year
adjustments were challenged in the two cases: the FY 2003 threshold in this case, and the FY
2004, FY 2005, and FY 2006 thresholds in District Hospital Partners. See id. at 167. In that
case, “[t]he question the Board certified was whether the ‘various elements used to project the
outlier thresholds were arbitrary and capricious.’” Id. at 168 (citing Complaint Ex. A at 1). That
court determined the plaintiffs’ request did not encompass any decisions not to adjust the
thresholds after the fact—which necessarily occurred after the setting of the fixed loss thresholds
for the applicable years. See id. In this case, however, Plaintiffs’ challenge encompassed a
challenge to the validity of the 2003 regulation, which promulgated changes to the outlier
payment regulations and encompassed the decision not to lower the FY 2003 fixed loss threshold
after the fact. The two cases, therefore, are distinguishable. Accordingly, the Court concludes that
Plaintiffs’ challenge to the June 9, 2003, decision to forego a mid-year adjustment of the FY
2003 fixed loss threshold was within the scope of the questions approved by the PRRB and,
therefore, that it is within the scope of this Court’s subject matter jurisdiction. The Court returns
to the merits of Plaintiffs’ claims regarding the agency’s decision not to undertake a FY 2003
mid-year adjustment below.
Furthermore, insofar as the Secretary’s argument that this Court lacks subject jurisdiction
is directed at “any other claims that were not properly channeled through the PRRB,” that
54
argument is unavailing as well. Def.’s Mot. at 30. Not only has the Secretary failed to identify
any other claims in this action that were not presented to the PRRB, but the Court concludes,
because it must assure itself of jurisdiction over the claims in this action, that each was
adequately presented to the PRRB. As explained above, the legal question presented to the
PRRB was framed broadly, referring to each of the revisions to the outlier payment regulations
and each of the fixed loss threshold regulations at issue for the payment years in question—that
is, each of the regulations challenged in this action. See PRRB Rs. at 29. Accordingly, Plaintiffs
have complied with the jurisdictional requirements of the Medicare Act, and the Court has
subject matter jurisdiction over each of the claims in this action.
Third, and finally, the Secretary argues that claims aimed at increasing Plaintiffs’ outlier
payments rather than at correcting an identifiable error would be barred by sovereign immunity.
It is true that Plaintiffs cannot bring a claim under the Medicare Act, which incorporates the
APA’s standard of review, in order to simply increase Plaintiffs’ payments that they receive from
the government. However, Plaintiffs are not doing so: they have brought claims identifying what
they argue are legal errors in each of the 14 rulemakings at issue in this case. That is enough to
fall within the government’s waiver of sovereign immunity.
Having concluded that the Court has subject matter jurisdiction over the claims in this
action, the Court turns to those claims—the legal errors Plaintiffs purport to identify with respect
to the 14 regulations in question—in the remainder of this Memorandum Opinion. However,
before addressing the individual claims, in light of the multitude of arguments that Plaintiffs
present with respect to the 14 regulations at issue in this case and the variety of arguments that
the Secretary presents in response, the Court reviews several fundamental principles regarding
the parameters of its review.
55
C. Framing Issues
Essentially, Plaintiffs present two sets of challenges in this case. First, Plaintiffs argue
that certain of the challenged regulations—specifically the outlier payment regulations from
1988 and 1994 and all of the fixed loss threshold regulations at issue—are at odds with the
language of the Medicare Act. 23 Plaintiffs refer to this set of arguments as their Chevron
challenge—under Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837
(1984). See Pls.’ Mot. at 25. Second, Plaintiffs argue that each of the regulations at issue in this
case was arbitrary and capricious. While Plaintiffs’ claims pertain to 14 different regulations over
the course of a thirty year period, there are several legal questions that recur in the parties’
briefing. In the interest of clarity, the Court first addresses several recurring legal issues before
addressing Plaintiffs’ individual challenges. The resolution of these issues sets the stage for the
Court’s discussion of Plaintiffs’ specific claims below.
Of these several framing questions, the Court first addresses Defendant’s argument that
Plaintiffs are barred from raising various arguments by failing to raise them in comments before
the agency at the time of the various rulemakings or, at a minimum, being able to point to other
parties who raised those arguments before the agency at the proper time. The Court disagrees.
The Medicare Act allows judicial review of legal issues pertaining to regulations pursuant to the
scheme described above: by filing timely appeals of payment determinations with the PRRB and
seeking judicial review on legal issues outside the scope of the PRRB’s authority. See 42 U.S.C.
§ 1395oo(a)(3); id. § 1395oo(f)(1). Plaintiffs have done so here. Indeed, Defendant does not
contest the timeliness of Plaintiffs’ challenges to the payment determinations from the relevant
23
Plaintiffs also argue that the failure to engage in reconciliation in FY 2004 through FY 2007
also violates the clear language of the statute. As discussed further below, the Court understands
this argument as linked to Plaintiffs’ arguments regarding the fixed loss threshold regulations for
those years.
56
fiscal years. Nor does Defendant contend that a challenge to those payment determination cannot
be the appropriate vehicle for challenging the regulations on which those payment
determinations depend. However, Defendant argues, nonetheless, that various arguments by
Plaintiffs are barred because they were never placed before the agency during the various
rulemaking proceedings subject to challenge in this action. In support of this proposition, they
cite to various cases that stand for the proposition that arguments must be raised before an
agency before they can be raised in court. See, e.g., Nuclear Energy Inst., Inc., v. EPA, 373 F.3d
1251, 1298 (D.C. Cir. 2004). But those cases do not tell the whole story. Even where a party has
waived its opportunity to pursue facial review of a regulation by failing to comment during a
rulemaking proceeding, 24 such a party can raise its arguments when the agency applies the rule.
See Koretoff v. Vilsack, 707 F.3d 394, 399 (D.C. Cir. 2013) (citing Murphy Exploration &
Production Co. v. U.S. Department of Interior, 270 F.3d 957, 958 (D.C. Cir. 2001)). The
Secretary responds, in turn, that this is a facial challenge because Plaintiffs are arguing that the
regulations in question cannot be validly applied in calculating any hospital’s payments—not
simply in calculating payments due to Plaintiffs. See Def.’s Reply at 17. For that proposition, the
Secretary looks to the oft-repeated standard for success on a facial challenge: “‘To prevail in
such a facial challenge, [a party] ‘must establish that no set of circumstances exists under which
the [regulations] would be valid.’” Ass’n of Private Sector Colleges & Universities v. Duncan,
681 F.3d 427, 442 (D.C. Cir. 2012) (citation omitted). However, in doing so, the Secretary has
transformed the standard for success on a facial challenge into the definition of a facial
24
The Court notes that, even in the facial challenge posture, “the waiver rule would not bar a
facial challenge if the agency has actually addressed the issue, either sua sponte or at the behest
of another party.” Koretoff, 707 F.3d at 400 n.3 (Williams, J., concurring) (citing Ohio v. EPA,
997 F.2d 1520, 1529 (D.C. Cir. 1993)).
57
challenge. They are not the same. The Secretary has not pointed to any authority suggesting that,
just because a plaintiff argues that a regulation is invalid, such a plaintiff has waived any
arguments not raised in a prior rulemaking proceeding. To the contrary, a series of cases from the
D.C. Circuit Court of Appeals indicate that a party may challenge the very validity of a
regulation when that regulation is applied without waiving arguments that were not raised before
the agency in the underlying rulemaking proceedings. See Weaver v. Fed. Motor Carrier Safety
Admin., 744 F.3d 142, 145 (D.C. Cir. 2014) (citing cases where the Court of Appeals “considered
the validity of rules that an agency applied” “despite want of a prior timely attack” to the
regulations in question) (emphasis added); see also Natural Res. Def. Council v. EPA, 513 F.3d
257, 260 (D.C. Cir. 2008) (describing “established doctrine that parties claiming substantive
invalidity of a rule for which direct statutory assault is time-barred are nonetheless free to raise
their claims in actions against agency decisions applying the earlier rule”) (emphasis in original).
That is precisely what Plaintiffs have done here. They have raised a timely challenge to the
outlier payments they received for FY 1997 through FY 2007, which entailed applying the 14
regulations subject to challenge in this action. Accordingly, they may raise arguments
challenging the validity of those regulations regardless of whether those arguments were raised
before the agency in the challenged rulemakings.
However, this conclusion is not the end of the story. Just because Plaintiffs are not barred
from raising at this time arguments that were not previously raised before the agency does not
mean that they will prevail on them. Indeed, whether comments have been presented to the
agency matters in applying the arbitrary and capricious standard of review. In addition to other
potential bases for determining that an agency action was arbitrary or capricious, discussed
further below, this Court cannot “uphold agency action if it fails to consider ‘significant and
58
viable and obvious alternatives.’” Dist. Hosp. Partners, 786 F.3d at 59 (quoting Nat’l Shooting
Sports Found., Inc. v. Jones, 716 F.3d 200, 215 (D.C. Cir. 2013) (internal quotation marks
omitted)). Determining whether comments have been presented to the agency regarding a
specific challenge is thus part—but not all—of the calculus entailed in determining whether the
agency has failed to consider “significant and viable and obvious alternatives.” The Court will
apply that standard below, as required, in resolving Plaintiffs’ arbitrary and capricious claims.
One particular application of that rule is worth explaining at this point because of its
general relevance to this case. Many of Plaintiffs’ arguments are dependent, in some form, on the
differences between the draft interim final rule sent to the Office of Management and Budget in
2003 and the subsequent rulemakings undertaken by the agency. However, the Court of Appeals
has determined that Plaintiffs cannot “[r]ely[] on the OMB draft rule to impugn the 2004
rulemaking.” Dist. Hosp. Partners, 786 F.3d at 58. Because “the OMB draft was never ‘on the
books’ in the first place”—that is to say, never published in the Federal Register—the Court of
Appeals concluded that the agency had no obligation to explain its departure from the provisions
of that internal draft. Id. That conclusion is wholly applicable here.
Furthermore, as the Court of Appeals recently clarified in District Hospital Partners,
contrary to Plaintiffs’ arguments here, the agency does not have an obligation to use the “best
available data” above and beyond the requirements of arbitrary and capricious review. See id. at
56 (rejecting argument that Gas Appliance Manufacturers Ass’n, Inc. v. DOE, 998 F.2d 1041
(D.C. Cir. 1993), “imposed a generic obligation on agencies to always use the best available
data”).
Next, the Court turns to the basis for its review of the various agency actions challenged
here. A fundamental principle of judicial review of agency actions is that such review is based
59
on—and limited to—the record before the agency at the time of the action. See Walter O.
Boswell Mem’l Hosp., 749 F.2d at 792 (“If a court is to review an agency’s action fairly, it should
have before it neither more nor less information than did the agency when it made its decision.”);
see also 5 U.S.C. § 706 (for review under the APA, “the court shall review the whole record or
those parts of it cited by a party”). The Supreme Court has cautioned that the record must be
limited to that which existed at the time of the decision. See Citizens to Preserve Overton Park,
401 U.S. at 420 (“[R]eview is to be based on the full administrative record that was before the
Secretary at the time he made his decision.”) (emphasis added). Accordingly, information that
was not yet before the agency at the time of a particular agency decision cannot provide the
ammunition for challenging that decision.
Here, Plaintiffs are challenging 14 discrete regulations. Only the information that was
before the agency at the time of each of those individual rulemakings can be used to challenge
that action. In evaluating each rulemaking, the Court must exclude all information that pertains
to events after that rulemaking, including information in the administrative records for
subsequent rulemakings. For example, in evaluating the FY 2004 fixed loss threshold
rulemaking, anything outside of the administrative record for that rulemaking—including
information in the administrative records for subsequent fixed loss threshold rulemakings—
cannot be considered. The same result applies to the Court’s evaluation of each of the three
outlier payment rulemakings and the 11 fixed loss threshold rulemakings challenged in this
action.
Furthermore, notwithstanding Plaintiffs’ suggestions to the contrary, the mere fact that a
subsequent rulemaking relied on a prior rulemaking does not mean that the earlier rulemaking
becomes open to challenge as a result of information before the agency at the time of the
60
subsequent rulemaking. See Biggerstaff v. FCC, 511 F.3d 178, 184 (D.C. Cir. 2007) (describing
reopening doctrine). Plaintiffs are correct that the issue directly before the Court of Appeals in
Biggerstaff was whether the petitioners in that case were time barred from challenging an earlier
agency action, which those petitioners claimed was encompassed in a later rulemaking. As
Plaintiffs argue, the issue of timeliness is not before this Court because Plaintiffs’ challenges to
each of the 14 regulations on which the challenged payment determinations rely are timely, as
explained above. However, the Court concludes that the reasoning in Biggerstaff, and the
associated line of reopening cases, is equally applicable here. In Biggerstaff, the Court of
Appeals concluded that, “for the court to examine the merits of his contention, [plaintiff] must
demonstrate that in the [subsequent] rulemaking the Commission reopened consideration of its
authority to adopt the [earlier-adopted] exemption, for otherwise his challenge is untimely.”
Biggerstaff, 511 F.3d at 185. The Court of Appeals also concluded that the agency’s “intention to
initiate a reopening must be clear from the administrative record.” Id. It is sensible to apply this
standard to determine whether actions taken in an earlier rulemaking can be evaluated in light of
the information in front of the agency at a later date. If the agency effectively reopened an earlier
rulemaking, such earlier decisions can be subject to challenge as a result of the later rulemaking.
Similarly, it is sensible that decisions made earlier can be challenged based on information
before the agency at the time of a later rulemaking only if the later rulemaking actually
considered and adopted, anew, the results of the earlier rulemaking—and only with respect to
applications of that decision after the later rulemaking. 25
25
To be clear, a subsequent re-adoption of a prior rule could not make that earlier rule
retroactively invalid as of the time it was originally adopted; but it could potentially make the re-
opened and re-adopted rule invalid as of the later adoption based on judicial evaluation of the
information before an agency at the time of the later adoption.
61
Despite Plaintiffs’ suggestions that the reopening line of cases is inapposite, Plaintiffs
have pointed to no authority suggesting an exception to the general rule that judicial review of an
agency action is limited to the record before the agency at the time of the agency action.
Accordingly, in the analysis below, the Court will evaluate the decisions made in each
rulemaking by analyzing the record before the agency at the time of each of those rulemakings.
Only in the limited circumstance where Plaintiffs have shown that one of the rules challenged
has been reopened and adopted through a later rulemaking will the Court review the later
adopted rule pursuant to the record before the agency at that later time, as well. 26
A final word is necessary regarding Plaintiffs’ framing of their challenges. While
Plaintiffs frame their Chevron arguments as a separate set of challenges, distinct from their
arbitrary and capricious arguments, Chevron does not provide an independent basis for reviewing
the agency’s actions. See Ass’n of Private Sector Colleges & Universities, 681 F.3d at 441.
Pursuant to the APA standard of review that is adopted by the Medicare Act, the Court reviews
agency actions, as relevant here, to determine whether they are “arbitrary, capricious, an abuse of
discretion, or otherwise not in accordance with law,” 5 U.S.C. § 706(2)(a), or “in excess of
statutory jurisdiction, authority, or limitations, or short of statutory right,” id. § 706(2)(C). 27
Chevron provides the two-step rubric for determining whether the agency actions were “not in
26
Somewhat analogously, Plaintiffs argue in several contexts that, in setting the fixed loss
threshold for a given year, the agency was aware of the relationship between actual outlier
payments from a previous year and previous projections as to outlier payments used in setting
the fixed loss threshold for that previous year. Although that claims recurs several times, it
requires a fact-specific inquiry—particular to the specific record before the agency at the time of
the respective challenged regulations. Accordingly, the Court will consider those arguments in
the context of Plaintiffs’ specific claims discussed below.
27
Notably, Plaintiffs never once cite the relevant statutory provisions that establish the
parameters for judicial review over agency action, which they seek in this case.
62
accordance with law” or “in excess of statutory jurisdiction, authority, or limitations, or short of
statutory right.” 28 See Ass’n of Private Sector Colleges & Universities, 681 F.3d at 441; Military
Toxics Project v. EPA, 146 F.3d 948, 954 (D.C. Cir. 1998). Accordingly, this Court applies the
analytic framework of Chevron in the context of Plaintiffs’ claims that the rulemakings at issue in
this case are inconsistent with the provisions of the Medicare Act. 29
Finally, even though Chevron does not provide an independent basis for review, in the
interest of clarity, the Court will address Plaintiffs’ challenges according to the manner in which
Plaintiffs organized them: first, a set of statutory challenges invoking Chevron with respect to
various regulations and, then, a series of claims that each of the regulations at issue in this case
was arbitrary and capricious. It is to the challenges invoking Chevron that the Court turns first.
D. Plaintiffs’ Chevron Arguments
Plaintiffs argue that the outlier payment model as a whole in effect for the fiscal years in
question was inconsistent with the outlier payment provisions of the Medicare Act, 42 U.S.C.
§ 1395ww(5)(A). Specifically, Plaintiffs claim that the payment model was inconsistent with
what they claim was the statute’s mandate to reimburse only for high cost cases because, they
claim, the agency actually reimbursed hospitals for high charge cases. Pls.’ Mot. at 25.
Defendant argues that the agency’s actions were a reasonable means of implementing the
28
Under the familiar Chevron standard, “[i]f the court determines ‘Congress has directly spoken
to the precise question at issue,’ and ‘the intent of Congress is clear, that is the end of the
matter.’” S. Carolina Pub. Serv. Auth. v. FERC, 762 F.3d 41, 54 (D.C. Cir. 2014) (quoting
Chevron, 467 U.S. at 842). “If, however, ‘the statute is silent or ambiguous with respect to the
specific issue,’ then the court must determine ‘whether the agency’s answer is based on a
permissible construction of the statute.’” Id. (quoting Chevron, 467 U.S. at 843).
29
The Courts notes, as well, that Plaintiffs have pointed to nothing that suggests that this Court’s
analysis of their so-called Chevron challenges, including the Chevron step two determination of
whether an agency’s interpretation of a statute is permissible, is not limited to the record before
the agency the time of the respective rulemakings.
63
statutory provisions. Defendant also argues that Plaintiffs’ challenges are barred because they
were not presented to the agency previously. With respect to the latter argument, the Court
concludes that, for the reasons stated above, Plaintiffs may raise these arguments here even if
they were not presented in comments to the agency because this is a timely challenge to the
regulations as applied in outlier payment determinations challenged in this action. See Koretoff,
707 F.3d at 399 (arguments not raised before agency may be raised in timely challenge to
application of rule). With respect to the former argument, the Court agrees with Defendant: each
of the agency actions challenged as inconsistent with the statute entailed a reasonable,
interpretation of the statute. The Court reviews Plaintiffs’ arguments to the contrary here.
1. Challenge to Outlier Payments for FY 1997 through FY 2003
With respect to FY 1997 through FY 2003, the gravamen of Plaintiffs’ claim appears to
be that the agency made outlier payments for these years where the hospitals’ actual costs were
below the “cut off point,” 42 U.S.C. § 1395ww(5)(A)(iii), and that such payments are barred by
the language of the statute. There are several fatal flaws with Plaintiffs’ argument. First,
Plaintiffs persist in referring to the outlier payment regulations and the fixed loss threshold
rulemakings applicable in these fiscal years as a bundle. Plaintiffs do not pinpoint explicit
decisions in the nine agency rulemakings at play for this set of fiscal years—the revisions to the
outlier payment regulations promulgated in 1988 and 1994, as well the fixed loss threshold
rulemakings for these seven fiscal years—that are inconsistent with the requirements of the
statute. As the Court has stated previously, Plaintiffs’ challenges must be tethered to specific
agency actions. See Banner Health I, 797 F. Supp. 2d at 106. Plaintiffs’ failure to do so
adequately at this point would be enough, without more, to make Plaintiffs’ statutory challenge
fail. However, in the interest of thoroughness, the Court considers whether the underlying
64
rulemakings are inconsistent with the statute. For the reasons stated below, the Court concludes
that they are not inconsistent because the statute does not require what Plaintiffs claim it
requires.
The Court begins, as it must, with the language of the statute itself. The statute requires
that the agency make payments, upon request, “in any case where charges, adjusted to cost”
exceed a certain cutoff point. 42 U.S.C. § 1395ww(5)(A)(ii). As a result of amendments to the
statute enacted in 1993, that cutoff point—for discharges after October 1, 1994, including all the
years at issue in this action—is the “sum of the applicable DRG prospective payment rate … plus
a fixed dollar amount determined by the secretary.” 30 Id. That fixed dollar amount is the amount
that has been referred to throughout this Memorandum Opinion as the fixed loss threshold. As a
reminder, the DRG prospective payment rate is the rate set for a particular year for a particular
category of medical diagnoses. The statute does not directly tell the agency how to set the fixed
loss threshold. However, the statute does require that “[t]he total amount of the additional
payments made under this subparagraph for discharges in a fiscal year 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.” Id. § 1395ww(5)(A)(iv). The D.C.
Circuit Court of Appeals has determined that the agency complies with this requirement by
“selecting outlier thresholds that, ‘when tested against historical data, will likely produce
aggregate outlier payments totaling between five and six percent of projected … DRG-related
payments.’” Dist. Hosp. Partners, 786 F.3d at 51 (quoting Cnty. of Los Angeles, 192 F.3d at
30
As explained above, the calculus also includes other payments not subject to this challenge—
“any amount payable under subparagraphs (B) and (F)”—but leaving those payments to the side
simplifies the narrative but does not subtract anything from the relevant analysis. 42 U.S.C.
§ 1395ww(5)(A)(ii); see Dist. Hosp. Partners, 786 F.3d at 50 n.1.
65
1013). Finally, the statute directs the agency that the “amount of such additional payment …
shall … approximate the marginal cost of care beyond the cutoff point applicable.” Id.
§ 1395ww(5)(A)(iii).
With these remarkably few words, Congress established the framework for the outlier
payment scheme, leaving the details to be fleshed out by the agency. Cf. Cnty of L.A., 192 F.3d at
1016 (“Where, as here, Congress enacts an ambiguous provision within a statute entrusted to the
agency’s expertise, it has ‘implicitly delegated to the agency the power to fill those gaps.’”)
(quoting National Fuel Gas Supply Corp. v. FERC, 811 F.2d 1563, 1569 (D.C. Cir. 1987)).
Specifically, the statute establishes the framework for determining both when outlier payments
should be made and what amounts those payments should be. But the details necessary to
implement the statutory framework and to fulfill the statutory mandate are left to the agency.
Plaintiffs argue that the agency wrongly based outlier payments on cases with high
charges rather than high costs. See, e.g., Pls.’ Mot. at 25. However, notwithstanding Plaintiffs’
cut-and-paste citations of the statute, they ignore key language in the relevant statutory
provision. The statute specifies that outlier payments are available where “charges, adjusted to
cost” exceed the cutoff point. 42 U.S.C. § 1395ww(5)(A)(ii). In other words, the statute requires
the agency to make the determination of when an outlier payment should be made based on
“charges, adjusted to cost,” not based on hospital costs in the abstract. The statute does not
specify how to adjust charges to cost, but the statute clearly requires the agency take charges and
adjust them to cost, in some manner, in order to make that determination. In 1988, in a
rulemaking challenged in this action, the agency determined that it would adjust charges to cost
by using hospital-specific cost-to-charge ratios. 53 Fed. Reg. at 38,503. These cost-to-charge
ratios would be derived by using historical data, specifically using the latest settled cost reports.
66
Id. at 38,529. The agency also concluded that, for hospitals where the data indicated extremely
high or extremely low cost-to-charge ratios, the agency instead would use the relevant statewide
average cost-to-charge ratio to adjust charges to cost for those hospitals. Id. at 38,507-08.
Through that rulemaking, the agency amended the applicable regulations, codified at 42 C.F.R.
§ 412.84(h). These decisions were not revisited, nor were the relevant provisions of the
regulations amended, until later proceedings in 2003, discussed elsewhere in this Memorandum
Opinion. Therefore, these provisions were applicable for the fixed loss threshold rulemakings for
FY 1997 through FY 2003.
None of these conclusions are inconsistent with the statute. The statute plainly indicates
that charges adjusted to cost should be used to determine whether an outlier payment is
applicable for an individual case. See 42 U.S.C. § 1395ww(5)(A)(ii). The statute is just as plainly
ambiguous regarding the method of adjusting charges to cost. Because the statute is ambiguous,
the Court “defers to the agency’s interpretation as long as it is based on a permissible
construction of the statute.” Chevron, 467 U.S. at 843. “The question for a reviewing court is
whether in doing so the agency has acted reasonably and thus has ‘stayed within the bounds of
its statutory authority.’” Natural Res. Def. Council v. EPA, 777 F.3d 456, 463 (D.C. Cir. 2014)
(quoting City of Arlington v. FCC, 133 S. Ct. 1863, 1868 (2013)). The Court concludes that each
decision in the 1988 regulation regarding the adjustment of charges to cost was reasonable. 31
The agency’s decision to adjust charges to cost by looking at hospital-specific data was
reasonable. See 53 Fed. Reg. at 38,503 (describing conclusion that “use of hospital-specific cost-
to-charge ratios should greatly enhance the accuracy with which outlier cases are identified and
31
The Court addresses Plaintiffs’ specific arguments that the agency did not adequately explain
certain decisions in analyzing, below, Plaintiffs’ claims that each regulation at issue in this case
was arbitrary and capricious.
67
outlier payments are computed, since there is wide variation among hospitals in these cost-to-
charge ratios”). So, too, was the agency’s decision to adjust charges to cost using historic cost-to-
charge ratios that are based on the latest settled cost reports. See 53 Fed. Reg. at 38,507
(explaining decision to use this dataset). As explained further below, just because the agency
later concluded—in 2003—that using somewhat more recent data (the tentatively settled cost
reports) was more sensible in light of the agency’s experience does not make the earlier decision
to use settled cost reports unreasonable. See 68 Fed. Reg. at 34,497-99. Similarly, the agency’s
decision to use statewide average data for hospitals with extremely high or low cost-to-charge
ratios was also reasonable. See 53 Fed. Reg. at 38,507-08 (explaining agency’s assessment that
“that ratios falling outside this range are unreasonable and are probably due to faulty data
reporting or entry”). Just because the agency later concluded—in 2003—that it would be more
sensible in light of the agency’s experience to use hospital-specific charge ratios for hospitals
with low cost-to-charge ratios rather than statewide averages, see 68 Fed. Reg. at 34,497-99,
does not mean that the decision to use statewide averages in 1988 was an unreasonable
construction of the statute. The Court notes that the agency never revisited these several
decisions until 2003 when the agency ultimately revised the outlier payment regulations.
Accordingly, the Court has no occasion to evaluate the reasoning behind these decisions in light
of events that occurred in those intervening years.
The same considerations apply with respect to the “amount of such additional payment”
under section 1395ww(5)(A)(iii). The statute requires that that amount “approximate the
marginal cost of care beyond the cutoff point applicable under clause (i) or clause (ii).” 42
U.S.C. § 1395ww(5)(A)(iii). In a 1983 regulation not challenged in this action, the agency
determined that the amount of a cost outlier payment would be determined by applying a
68
marginal cost factor to the difference between charges, adjusted to cost, and the applicable
threshold. 32 See 48 Fed. Reg. 39,752, 39,777 (Sept. 1, 1983). The Court notes that it does not
appear that the agency actually reconsidered the decision to use cost-adjusted charges to
determine the amount of the outlier payment in promulgating the 1988 revisions to the outlier
payment. See 53 Fed. Reg. at 38,502 (describing the consideration of the level of the marginal
cost factor without reconsidering the use of cost-adjusted charges). In any event, Court concludes
that it is reasonable to determine the amount of an outlier payment based on cost-adjusted
charges derived by the same methodology of cost-adjustment as used to determine whether a
case qualifies for an outlier payment.
It is important that this clause requiring that the “amount of such additional payment …
shall approximate the marginal cost of care beyond the cutoff point applicable” is applicable both
to day outliers and to cost outliers. 42 U.S.C. § 1395ww(5)(A)(iii). The language of this clause is
necessarily capacious enough to accommodate both forms of outliers. 33 It is reasonable that the
agency treated “cost of care” similarly to the calculation of “charges, adjusted to cost” used to
determine whether a cost outlier payment was applicable for an individual case—in accordance
with the provisions regarding the outlier threshold just discussed. Pursuant to the regulations that
32
In the 1983 rulemaking, the agency explained why it was necessary to use a marginal cost
factor: “Marginal cost is the change in total cost associated with a one unit change in output. Due
to the presence of fixed costs, the marginal cost of care is generally less than the average cost.”
48 Fed. Reg. at 39,776. The use of a marginal cost factor to transform the average costs above
the threshold into the marginal costs above that threshold is not in dispute. Nor are the individual
marginal cost factors that the agency used in the fiscal years relevant to this dispute.
33
“For day outliers, an additional per diem payment is made for each covered day of care beyond
the length of stay threshold.” 53 Fed. Reg. at 38,502; see 42 C.F.R. § 412.82. “The per diem
payment is equal to 60 percent of the average per diem Federal rate for the DRG, which is
calculated by dividing the wage-adjusted Federal rate for the DRG by the geometric mean length
of stay for the DRG.” 53 Fed. Reg. at 38,502. The applicable percentage was set annually. 42
C.F.R. § 412.82. Day outliers are not at issue in this action.
69
the agency promulgated, the agency adjusted charges to cost to determine whether those cost-
adjusted charges were above the applicable threshold, and then made a payment based on the
amount by which the cost-adjusted charges exceeded that threshold. See 53 Fed. Reg. at 38,503.
Although the statute does not refer to “cost of care” in clause (iii) as being derived from
“charges, adjusted to cost,” as it did in clause (i) regarding the determination of whether an
outlier payment is applicable, the agency’s interpretation that cost-adjusted charges should be
used to determine the amount of the cost outlier payment—just as cost-adjusted charges were
used to determine whether a case exceeds the threshold for payment—is wholly reasonable. See
Nat’l Credit Union Admin. v. First Nat. Bank & Trust Co., 522 U.S. 479, 501 (1998) (describing
“established canon of construction that similar language contained within the same section of a
statute must be accorded a consistent meaning”); see also Janko v. Gates, 741 F.3d 136, 141
(D.C. Cir. 2014) cert. denied, 135 S. Ct. 1530 (2015). So, too, is the agency’s determination that
it should use the same methodology for cost-adjustment in determining the amount of the outlier
payment as in determining the applicability of the outlier payment. Not only does consistency in
this context have the virtue of simplicity, it is also plainly reasonable.
In District Hospital Partners, the D.C. Circuit Court of Appeals used an example to
explain how outlier payments were generated for FY 2004 through FY 2006 and to explain the
relationship among fixed loss thresholds, cost-adjusted charges, and outlier payments for those
fiscal years. 786 F.3d at 50-51 (citing 62 Fed. Reg. at 45,997). Similarly, an example may be
helpful in explaining the relationship between the determination whether a specific case qualifies
for an outlier payment and the determination of the amount of any such outlier payment—with
70
respect to the years at issue in this action. 34 Assume that the “cutoff point” for a certain DRG
category is $20,000. Pursuant to the regulations, the agency would adjust charges to cost based
on the applicable cost-to-charge ratio. If that cost-adjusted charge was above $20,000, a payment
would be available. If the charge for an individual case were $40,000 and the applicable cost-to-
charge ratio were 0.75 (or 75 percent), the cost-adjusted charge would be $30,000. Because that
amount is above $20,000, a payment would be available. The amount of the payment would be
the difference between the cost-adjusted charge ($30,000 in this example) and the threshold
($20,000)—after applying a marginal cost factor. If the marginal cost factor were 0.80, the
payment would be $8,000, or 80 percent of the difference between the cost-adjusted charge and
the threshold. It is reasonable to use the same the cost-adjusted charge—here a $40,000 charge
adjusted to a $30,000 estimated cost—to determine the amount of the payment as used to
determine whether a payment was warranted in the first place. Accordingly, the court concludes
that using cost-adjusted charges, pursuant to the methodology outlined above, to determine the
amount of the payment is a reasonable interpretation of the statutory requirement that the
“amount of the payment … shall … approximate the marginal cost of care beyond the cutoff
point applicable.” 42 U.S.C. § 1395ww(5)(A)(iii). In other words, the 1988 rulemaking was a
reasonable construction of the statute and is therefore consistent with the statute’s terms.
Turning to the 1994 rulemaking, only a brief discussion is necessary. In 1993, Congress
introduced the fixed loss threshold as the method of determining the applicable cutoff point used
to determine whether an outlier payment is available for a given case and to determine the
34
This analysis is equally applicable with respect to the statutory scheme extant in 1988 and with
respect to the statutory scheme as revised in 1993, when the “fixed loss threshold” was
introduced. For the purposes of this example, it is not necessary to review the calculation of the
threshold itself under those two schemes.
71
amount of the outlier payments for such a case. 35 See 59 Fed. Reg. at 45,368. The following year,
the agency revised the outlier payment regulations to use the altered methodology specified by
statute for determining when payments would be available and the amounts of such payments.
See id. The changes to the outlier payment regulations in this rulemaking were limited to
implementing the 1993 statutory changes. See id. Although Plaintiffs name the 1994 rulemaking
as a source of the inconsistency between the statute and the regulations applicable for fiscal years
1997 through 2003, Plaintiffs point to nothing about the 1994 rulemaking that would create that
inconsistency. Perhaps Plaintiffs mean to suggest that the 1994 regulation implicitly re-adopted
the decisions promulgated in the 1988 regulation—although they do not identify anything in the
record suggesting that the agency had done so. Moreover, Plaintiffs do not point to anything that
would make those 1988 regulations unreasonable in 1994—but would not have made them
unreasonable in 1988. Accordingly, just as the Court concluded that the 1988 revisions to the
outlier payment regulations were consistent with the Medicare Act, the Court concludes that the
1994 regulations are consistent with the Medicare Act, as amended in1993.
Finally, the Court turns to the seven fixed loss threshold rulemakings for FY 1997
through FY 2003—when the 1988 and 1994 regulations were operative. Critically, each of those
rulemakings took the formula for outlier payments established by the applicable outlier payment
regulations—as amended in 1988 and then in 1994—as a given. For each year, taking those
formulas as given, the agency calculated a fixed loss threshold that was predicted to generate
35
Prior to the 1993 statutory amendments, outlier payments were made where “charges, adjusted
to cost, exceed a fixed multiple of the applicable DRG prospective payment rate, or exceed such
other fixed dollar amount, whichever is greater.” 42 U.S.C. § 1395ww(5)(A)(ii). Going forward,
outlier payments would be made “exceed the sum of the applicable DRG prospective payment
rate … plus a fixed dollar amount determined by the Secretary.” Id. That fixed dollar amount is
known as the fixed loss threshold.
72
outlier payments that would be 5.1 percent of total DRG-based payments. See 61 Fed. Reg. at
46,228 (FY 1997 rulemaking); 62 Fed. Reg. at 46,040 (FY 1998 rulemaking); 63 Fed. Reg. at
41,008 (FY 1999 rulemaking); 64 Fed. Reg. at 41,546 (FY 2000 rulemaking); 65 Fed. Reg. at
47,113 (FY 2001 rulemaking); 66 Fed. Reg. at 39,941 (FY 2002 rulemaking); 67 Fed. Reg. at
50,124 (FY 2003 rulemaking). This approach was approved by the D.C. Circuit Court of Appeals
in County of Los Angeles, 192 F.3d at 1013. The agency is not required to adjust the fixed loss
thresholds in retrospect to ensure that actual outlier payments are between 5 and 6 percent of
actual DRG-based payments. Id. The agency is only required to select threshold that “when
tested against historical data, will likely produce aggregate outlier payments totaling between
five and six percent of projected … DRG-related payments.” Id. Although the D.C. Circuit did
not address this particular issue in County of Los Angeles, it is plainly reasonable to use the then-
applicable formula for generating outlier payments, which would be used to calculate the actual
outlier payments for these several fiscal years, to determine the fixed loss threshold for those
years.
Plaintiffs argue that outlier payments during FY 2003, in particular, were inconsistent
with the statute because the agency was aware of “turbo-charging” practices by certain hospitals.
See Pls.’ Mot. at 30. The only evidence of this awareness on which Plaintiffs rely is the draft
interim final rule sent to the Office of Management and Budget. See id. (citing A.R. at 4417.372,
4417.396). However, that draft was dated February 12, 2003—approximately six months after
the August 1, 2002, promulgation of final FY 2003 fixed loss threshold rule. See 67 Fed. Reg. at
50,124. Plaintiffs cannot rely on information that was only before the agency after the
promulgation of the FY 2003 rulemaking to impugn that rule. Moreover, like each of the other
fixed loss threshold rulemakings from FY 1997 through FY 2003, the FY 2003 rulemaking was
73
based on the established methodology for outlier payments applicable to that fiscal year, and the
agency did not reconsider the methodology in the FY 2003 rulemaking. Accordingly, the FY
2003 fixed loss threshold rulemaking is consistent with the statute just as its predecessors were.
Insofar as Plaintiffs are claiming that the failure to implement a mid-year adjustment to the FY
2003 fixed loss threshold was inconsistent with the statute, that claim fails as well. There is no
authority supporting Plaintiffs’ suggestion that a mid-year adjustment is required. To comply
with the statute, all that is required is for the agency to “select[] outlier thresholds that, ‘when
tested against historical data, will likely produce aggregate outlier payments totaling between
five and six percent of projected … DRG-related payments.’” Dist. Hosp. Partners, 786 F.3d at
51. The agency did so in selecting the FY 2003 fixed loss threshold on August 1, 2002.
Moreover, as explained in more detail below, in its final rule issued June 9, 2003, the agency
adequately explained it decision not to implement a mid-year adjustment to the FY 2003 fixed
loss threshold. See 68 Fed. Reg. at 34,505. The statute requires no more.
In sum, for the fiscal years between 1997 and 2003, the agency made outlier payments
based on the methodology established by the then-applicable outlier payment regulations, revised
in both 1988 and 1994; pursuant to that methodology, the fixed loss threshold chosen in an
annual fixed loss threshold rulemaking was a key input in determining the actual outlier
payments. As described above, each of those regulations is consistent with the statute. Plaintiffs
have identified no other ways in which the outlier payments for these years were inconsistent
with the statute. With respect to these seven fiscal years, Plaintiffs’ arguments amount to a claim
that the agency ultimately ended up making outlier payments to hospitals that had high charges
rather than hospitals that had high costs. As regrettable as that result may be, insofar as it is true,
that result does not show that the agency actions challenged here were incompatible with the
74
statute. 36 To the contrary, the Court concludes that outlier payment regulations applicable during
this period and the fixed loss thresholds established during this period were based on reasonable
constructions of the statute and, accordingly, the outlier payments during these fiscal years were
consistent with the statute, as well.
2. Failure to Conduct Reconciliation and to Account for Reconciliation
Next, Plaintiffs make a cursory argument that the agency violated the statute’s
requirement that outlier payments “shall … approximate the marginal cost of care beyond the
cutoff point applicable,” 42 U.S.C. § 1395ww(5)(A)(iii), by failing to conduct reconciliation or
to account for reconciliation in setting the fixed loss thresholds for FY 2004 through FY 2007.
This argument fails because nowhere does the statute require the agency to undertake
reconciliation. In 2003, the agency promulgated revisions to the outlier payment regulations that,
among other changes, stated that outlier payments would “become subject to adjustment when
hospitals’ cost reports coinciding with the discharge are settled.” 68 Fed. Reg. at 34,504
(codified at 42 C.F.R. § 412.84(i)(4)). The agency reasoned that, even with other changes to the
regulations promulgated that year, hospitals would be able to “manipulate the system to
maximize outlier payments” by “dramatically increas[ing] its charges by far above the rate-of-
increase in costs during any given year.” Id. at 34,501. To prevent manipulation of outlier
payments through dramatic charge increases, the agency introduced the possibility of
reconciliation. Through reconciliation, outlier payments that were already made—based on the
cost-to-charge ratios available at the time of the determination of those payments—could be
adjusted in light of the actual cost-to-charge ratios for the period in question. However, the
36
Insofar as Plaintiffs argue that the regulations on which the outlier payments relied were
arbitrary and capricious, the Court considers those arguments below.
75
agency’s regulations do not require reconciliation: they simply state that payments would
“become subject to adjustment,” rather than affirmatively requiring that they be adjusted. Id. at
34,504. Nor do the regulations indicate that reconciliation was necessary in order to ensure
consistency with the statute. Indeed, there is nothing to indicate that the failure to reconcile
would be inconsistent with the statutory requirement that outlier payments “shall … approximate
the marginal cost of care beyond the cutoff point applicable.” 42 U.S.C. § 1395ww(5)(A)(iii).
While reconciliation appears to be a plausible way to effectuate the statutory purpose in
combination with other changes made to the regulations, given the “turbo-charging” behaviors
that were discovered by the agency as of 2003, see 68 Fed. Reg. at 34,501, failing to conduct
reconciliation was not unreasonable and is therefore consistent with the statute.
Just as reconciliation itself is not mandated by the statute, neither does the statute
mandate that the agency account for reconciliation in setting the annual fixed loss threshold.
Because reconciliation became simply an option rather than a requirement, it would be irrational
to conclude that the statute actually required the agency to account for reconciliation explicitly in
calculating the fixed loss threshold. In addition, Plaintiffs wholly ignore the fact that, for FY
2004, the agency considered the impact of reconciliation in setting the fixed loss threshold. See
68 Fed. at 45,476-77. The agency’s methodology for that fiscal year integrated its estimate that
reconciliation would occur for approximately 50 hospitals—based on past hospital behavior and
based on the resources that would be available for reconciliation. Id. For FY 2005 through FY
2007, the agency reasonably concluded that, given other changes introduced by the 2003
revisions to the outlier payment regulations and given uncertainty regarding future behavior,
considering reconciliation in setting the fixed loss threshold was not warranted. See 69 Fed. Reg.
at 49,278 (FY 2005 rulemaking); 70 Fed. Reg. at 47,495 (FY 2006 rulemaking); 71 Fed. Reg. at
76
48,151 (FY 2007 rulemaking). In sum, the agency’s failure to reconcile and its failure to
explicitly account for reconciliation in fixed loss rule rulemakings for these years do not make
those rulemakings—or the outlier payments calculated pursuant to them—inconsistent with the
statute.
3. Challenge to Outlier Payment Determinations for FY 2004 through FY 2006
With respect to the fixed loss threshold rulemakings for FY 2004 through FY 2006,
Plaintiffs’ statutory argument is that the rulemakings did not comply with the requirement that
the agency choose “outlier thresholds that, ‘when tested against historical data, will likely
produce aggregate outlier payments totaling between five and six percent of projected … DRG-
related payments.’” Dist. Hosp. Partners, 786 F.3d at 51 (quoting Cnty. of Los Angeles, 192 F.3d
at 1013). Specifically, Plaintiffs argue that these three rulemakings did not comply with that
requirement because the agency did not adjust the cost-to-charge ratios used for those predictions
to account for continued declines in cost-to-charge ratios. See Pls.’ Mot. at 34-37. Accordingly,
Plaintiffs argue that it was predictable that the agency would choose fixed loss thresholds that
were too high and thus would result in outlier payments below the 5 to 6 percent range set by
section 1395ww(5)(A)(iv).
The Court notes that Plaintiffs repeat this claim in their argument that each of these three
fixed loss threshold rulemakings were arbitrary and capricious. The Court fully addresses the
agency’s responses to proposed alternatives regarding the adjustment of cost-to-charge ratios in
discussing Plaintiffs’ arguments that these fixed loss threshold rulemakings were arbitrary and
capricious. But in evaluating the consistency of these three annual rulemakings with the statute,
it is important that the agency thoroughly explained why it chose not to adjust the methodology
to account for any declines in cost-to-charge ratios:
77
We do not believe that it is necessary to make a specific adjustment to our
methodology for computing the outlier threshold to account for any decline in
cost-to-charge ratios in FY 2005, as the commenter has requested. We have
already taken into account the most significant factor in the decline in cost-to-
charge ratios, specifically, the change from using the most recent final settled cost
report to the most recent tentatively settled cost report. Furthermore, we strongly
prefer to employ actual data rather than projections in estimating the outlier
threshold because we employ actual data in updating charges, themselves.
69 Fed. Reg. at 49,277. With respect to FY 2005, the agency concluded that other changes to the
outlier payment regulations and to the fixed loss threshold methodology adequately accounted
for the decline in cost-to-charge ratios such that an additional adjustment of the cost-to-charge
ratios was not necessary. As explained below, the agency adequately explained similar choices
for FY 2004 and FY 2006. See 68 Fed. Reg. at 45,476-77; 70 Fed. Reg. at 47,495. Moreover, it is
proper to ‘‘‘defer to the agency’s decision on how to balance the cost and complexity of a more
elaborate model against the oversimplification of a simpler model,’’’ West Virginia v. EPA, 362
F.3d 861, 868 (D.C. Cir. 2004) (quoting Small Refiner Lead Phase-Down Task Force v. EPA, 705
F.2d 506, 535 (D.C. Cir. 1983)). Therefore, the Court will defer to the agency’s choice to use
actual historical data for the cost-to-charge ratios rather than to add an unknown adjustment
factor. See Disp. Hosp. Partners, 973 F. Supp. 2d at 16. The Court concludes that the agency’s
decision to use the actual cost-to-charge ratios in order to set the FY 2004 through FY 2006 fixed
loss thresholds, rather than adjusting those ratios to account for possible continued declines, was
reasonable. Accordingly, in these rulemakings, the agency has fulfilled the requirement that it set
“outlier thresholds that, ‘when tested against historical data, will likely produce aggregate outlier
payments totaling between five and six percent of projected … DRG-related payments.’” Dist.
Hosp. Partners, 786 F.3d at 51 (quoting Cnty. of Los Angeles, 192 F.3d at 1013).
* * *
78
In sum, the Court concludes that each of Plaintiffs’ statutory arguments fails because the
regulations challenged are consistent with the requirements of the Medicare Act. Each regulation
promulgated by the agency represents a reasonable interpretation of the outlier provisions of the
Medicare Act. Therefore, the outlier payments determined pursuant to these regulations are
necessarily consistent with the statute, as well. The Court now turns to Plaintiffs’ arguments that
the regulations challenged here were arbitrary and capricious.
E. Plaintiffs’ Arbitrary and Capricious Arguments
In addition to Plaintiffs’ claim that the several agency actions challenged here are
inconsistent with the statute, Plaintiffs argue that each of the regulations challenged here—the
revisions to the outlier payment regulations and the annual fixed loss threshold rulemakings—
were arbitrary and capricious. 37 At the outset, the Court reiterates that the Court’s review is
limited to the individual administrative records for each rulemaking—that is, the information that
was before an agency at the time of the respective rulemakings—and it cannot invalidate a
rulemaking because it subsequently becomes clear that a rulemaking was unwise. The Court
reviews each of Plaintiffs’ numerous arguments and concludes that, with one exception, none are
successful.
1. Challenges to the 1988 Revisions to the Outlier Payment Regulations
Plaintiffs present three challenges to what they call the “Outlier Payment Model.” As
explained above, the outlier payment regulations—amended in 1988, 1994, and 2003—establish
the methodology for the determination of annual payments. Plaintiffs challenge two features of
37
However, the Court notes that Plaintiffs’ arguments do not address the 2003 amendments to
the outlier payment regulations—with the exception of Plaintiffs’ argument regarding a mid-year
adjustment to the FY 2003 fixed loss threshold—notwithstanding Plaintiffs’ statement that they
challenge each of the outlier payment and fixed loss threshold rulemakings subject to this action.
79
the outlier payment regulations that were introduced in 1988 and then eliminated in revisions to
the outlier payment regulations in 2003. First, they challenge the use of “older” data—that is, the
use of the latest settled cost reports—to establish the hospital-specific cost-to-charge ratios,
which are used to adjust charges to cost. 38 Second, Plaintiffs challenge the use of statewide
average cost-to-charge ratios in lieu of hospital-specific cost-to-charge ratios for hospitals with
extremely low cost-to-charge ratios. Third, Plaintiffs present a challenge related to the use of
statewide average data: they argue that the statewide averages were outdated and based on
incorrect data. The Court considers each argument and concludes that none are successful.
Use of Latest Settled Cost Reports
In 1988, the agency concluded that, because cost-to-charge ratios differed among
hospitals, it would be preferable to use hospital-specific cost-to-charge ratios rather than a single
nationwide estimate, as the agency has done before. See 53 Fed. Reg. at 38,503 (reasoning that
“[t]he use of hospital-specific cost-to-charge ratios should greatly enhance the accuracy with
which outlier cases are identified and outlier payments are computed, since there is wide
variation among hospitals in these cost-to-charge ratios.”). The agency was, thus, faced with the
choice of what data to use to compute those cost-to-charge ratios. The agency explained its
choice to use the latest settled cost reports: “[w]hile the latest filed cost report represents the
most current data, we have found that Medicare costs are generally overstated on the filed cost
report and are subsequently reduced as a result of audit.” Id. The agency concluded that the
greater accuracy of audited cost-data outweighed the time lag necessitated by using that audited
data. See id.
38
As explained above, the agency then determines whether this cost-adjusted charge exceeds the
applicable threshold. If so, a payment is made based on the amount by which this cost-adjusted
charge exceeds the threshold.
80
The agency’s choice to use the latest settled cost reports was codified in amendments to
the regulations governing outlier payments. 53 Fed. Reg. at 38,529 (codified at 42 C.F.R.
§ 412.84(h) (“The cost-to-charge ratio used to adjust covered charges is computed annually by
the intermediary for each hospital based on the latest available settled cost report for that hospital
and charge data for the same time period as that covered by the cost report.”). The agency did not
revisit this choice until 2003, when the agency ultimately decided to made a different choice—
using the latest tentatively settled cost reports (when they were more recently updated than the
settled cost reports). See 68 Fed. Reg. at 34,497-99. Plaintiffs challenge only the earlier choice,
and, as stated above, this Court’s review is limited to the record before the agency in 1988.
Nothing that occurred later or that was presented to the agency later—even if it ultimately
convinced the agency to make a different policy choice—can be ammunition for challenging the
1988 decision to use the latest settled cost reports. In that light, the Court reviews Plaintiffs’
several arguments in support of their claim that this decision was arbitrary and capricious.
Plaintiffs claim that the agency ignored the warning of commenters that using settled cost
report data—rather than more recent data—would introduce inaccuracies into the calculation of
outlier payments. See AR (1988 Rulemaking) 7156, 7387. However, the agency responded to this
concern, acknowledging that settled cost reports did not represent the most recent data but
explaining that the agency “found that Medicare costs are generally overstated on the filed cost
report and are subsequently reduced as a result of audit.” 53 Fed. Reg. at 38,503. Therefore, the
agency chose the older-but-audited data instead of the most recent raw data. See id.
81
Plaintiffs next argue that, in choosing to use the settled cost reports, the agency
disregarded comments indicating that charges were rising faster than costs. 39 However, each of
the comments to which Plaintiffs cite for this proposition were submitted after the promulgation
of the final 1988 outlier payment regulations. See Pls.’ Mot. at 41 (citing 58 Fed. Reg. at 46,347
(1993); A.R. (1993 amendments) 6590; A.R. (1993 amendments) 6685; A.R. (FY 2003) 4809;
A.R. (2003 amendments) 4390; A.R. (2003 amendments) 4392; A.R. (2003 amendments) 4399;
A.R. (2003 amendments) 4401; A.R. (2003 amendments) 4407; A.R. (2003 amendments)
4417.357; A.R. (FY 2005) 1032; A.R (FY 2006) 609). Accordingly, Plaintiffs may not rely on
these comments in challenging the 1988 regulation, and the Court may not rely on them in
determining whether that regulation was arbitrary or capricious.
Plaintiffs also argue that the choice to use settled cost reports was unreasonable because,
even using settled cost reports, the agency’s calculation would necessarily involve unaudited
data. Plaintiffs point out that calculating outlier payments involves adjusting current charges,
which are unaudited, by a ratio that is calculated from historic cost data, which has been audited,
and historic charge data, which has not been audited. Accordingly, Plaintiffs argue that it was
irrational to choose older data, which includes audited cost data, at the expense of the timeliness
of the data, when they would nonetheless be using some unaudited data. The Court first notes
that the Secretary had no need to respond to this concern because Plaintiffs do not point to
anything in the record suggesting that this concern was before the agency at the time of the 1988
rulemaking. Moreover, as the Secretary argues, it was not irrational to use a data source that
39
The gravamen of the claim is that, if charges are rising faster than costs, the actual cost-to-
charge ratios would necessarily be falling over time. In turn, that trend would imply that using an
older cost-to-charge ratio—all things being equal—would mean using a cost-to-charge ratio that
was higher than the actual cost-to-charge ratio for the relevant time period.
82
included some audited data and some unaudited data. In particular, the agency emphasized the
importance of using the audited cost data in light of its finding “that Medicare costs are generally
overstated on the filed cost report and are subsequently reduced as a result of audit.” 53 Fed.
Reg. at 38,503. Accordingly, it was reasonable to choose to use an older data source on account
of the fact that it included audited cost data even while the charge data used was not audited.
Next, Plaintiffs argue that the agency disregarded its own previous statements—in years
prior to adopting the hospital-specific cost-to-charge ratio methodology—that “using hospital-
specific [cost-to-charge ratios] came with the risks of introducing data with poor predictive
value.” Pls.’ Mot. at 42. However, the several statements that Plaintiffs cite do not show that the
agency’s 1988 decision was arbitrary or capricious.
The first set of statements that Plaintiffs identify is a pair of statements in the 1986
rulemaking, in which the agency declined to shift from using a national cost-to-charge ratio to
using hospital-specific cost-to-charge ratios, as a commenter had then suggested. See 51 Fed.
Reg. 31,454, 31,526 (1986). First, the agency noted that “[t]he use of hospital-specific cost-to-
charge ratios to compute outlier payments would require that they be frequently revised to
account for changes in the mix and scope of services provided.” Id. This statement reflects the
(undisputed) additional administrative burden entailed by calculating hospital-specific ratios.
Notwithstanding the administrative burden associated with the methodology, in 1988, the agency
ultimately chose to adopt hospital-specific cost-to-charge ratios because it concluded that they
would be more accurate than a single nationwide ratio. See 53 Fed. Reg. at 38,503. The agency’s
1986 statement does not reveal anything arbitrary or capricious about the 1988 decision. 40 Also,
40
Notably, Plaintiffs do not suggest it would be preferable to use a nationwide rate (and indeed
vociferously contest the use of statewide averages).
83
in the 1986 rulemaking, the agency noted that “[a]n additional source of inaccuracy could arise
from the fact that the hospital-specific cost-to-charge ratios would presumably be from
completed cost reports, which means that they would lag behind the current period for which
they would be used in computing outlier payments.” 51 Fed. Reg. at 31,526. In later adopting the
use of hospital-specific cost-to-charge ratios, the agency did not ignore this lag, it simply
embraced the tradeoff entailed: it concluded that (1) using older but more accurate settled cost
reports was preferable to using more recent unaudited cost reports, and (2) using hospital-
specific ratios was more accurate than using a nationwide cost-to-charge ratio. See 53 Fed. Reg.
at 38,503. Accordingly, the agency adequately responded to the concern that it itself had voiced
in the 1986 rulemaking.
The second statement that Plaintiffs identify is a statement in the 1987 rulemaking in
which the agency explained why it decided once again not to adopt hospital-specific cost-to-
charge ratios at that time. It is important to set this decision in its context. The decision not to
adopt hospital-specific cost-to-charge ratios in 1987 was part of a larger decision to delay
broader changes to the outlier program, which would have shifted the emphasis from day outliers
(patients with extremely long stays in the hospital) to cost outliers (extremely high cost cases),
until the agency could complete its analysis of the relevant factors. See 52 Fed. Reg. at 33,047-
48. The agency described its reasoning as follows:
While the use of hospital-specific cost-to-charge ratios may be more accurate for
purposes of computing cost outlier payments, there are a number of significant
administrative and data problems associated with using these ratios. For example,
estimating future outlier payments in order to establish appropriate outlier
thresholds becomes more of a problem, since hospital specific cost-to-charge
ratios used in the estimate would not be the same as those used for actually paying
outliers. In addition, major changes in PRICER software (the program used to
calculate each hospital’s payment per discharge) and the Medicare cost report
would be necessary in order for cost-to-charge ratios to be developed for payment
purposes.
84
52 Fed. Reg. at 33,048. In part, the agency was highlighting an administrative and technical
problem regarding transitioning to a new system. Just as with the administrative burdens
highlighted in the 1986 rulemaking, these administrative and technical details were not simply
brushed aside in the 1988 rulemaking; rather, the agency concluded that the administrative
burden was worthwhile and decided that it was necessary to overcome the technical difficulties
entailed in the transition. The agency also identified a specific challenge with using hospital-
specific data: that the ratios used in setting the thresholds for outlier payments, in advance,
would not be the same as those used to calculate the actual outlier payments. Id. In retrospect,
there is no doubt this phenomenon has been a challenge for the agency, and the agency
ultimately changed several policy choices in 2003 in light of its experience over the past three
decades. However, in deciding to use hospital-specific ratios in 1988, the agency adequately
addressed the issue it has raised the previous year. The agency did not conclude that using the
settled cost reports would be a perfect solution, but as described above, the agency reasonably
concluded that it would be better than the available alternatives. See 53 Fed. Reg. at 38,503; cf.
Dist. Hosp. Partners, 786 F.3d at 59 (agency must address “significant and viable and obvious
alternatives”).
The third statement that Plaintiffs identify is a statement in the 1988 rulemaking
promulgating changes to the outlier payment regulations, which Plaintiffs claim shows that the
agency did not adequately address problems with the decision to use hospital-specific cost-to-
charge ratios. See Pls.’ Mot. at 42-43 (citing 53 Fed. Reg. at 38,509). Because Plaintiffs’ claim
regarding this statement extracts it from its essential context, it is worth reproducing in full the
agency’s summary of the comment to which it was responding and the agency’s response:
85
Comment: Some commenters were concerned that the increased emphasis on cost
outliers in the proposed policy would provide an incentive for hospitals to
increase their charges and to manipulate their charge structures.
Response: Cost outliers are identified by, and the amount of cost outlier payment
determined by, comparing the charges for the case, adjusted by a cost-to-charge
ratio, to the cost outlier threshold. Since both the cost-to-charge ratio (whether
national or hospital-specific) and the threshold are constant for the payment
period, the payment received by the hospital can be increased by increasing
charges. In addition, hospitals can conceivably change their charge structures, just
as is the case at present, to maximize their outlier payments.
Although concern over this type of incentive is appropriate, we believe that there
are several factors that will mitigate its effects. First, increases in a hospital’s
overall charges relative to costs will be reflected in the cost-to-charge ratio
assigned to the hospital in the future. This is one of the strong arguments for the
use of hospital-specific cost-to-charge ratios. Second, many hospitals are
restricted in their ability to arbitrarily increase their charges by the fact that they
must deal with other third-party payers, some of which base their payments on
charges. Also, several States place restrictions on hospital charge increases. Third,
a general acceleration in hospital charge increases can be incorporated into the
setting of thresholds in future years, which would limit the potential benefit to
hospitals.
Fourth, outlier payments comprise a small percent of total hospital payments
under the prospective payment system, diluting the incentive for hospitals to
disrupt their operations by drastically and continually manipulating charges.
It must be pointed out that this incentive to manipulate charges is not new; in fact,
any measure of cost (including length of stay) that is based on an indicator that is
within the control of the provider provides an incentive to manipulate that
indicator. As previously stated, we will continue to investigate potential
improvements in the measurement of case level costs.
53 Fed. Reg. at 38,510. Most importantly, in this response, the agency appears to be responding
to a comment regarding a major shift to the outlier payment regime introduced in 1988 that is not
challenged in this action: a shift from a system focused on day outliers (extremely lengthy stays)
to cost outliers (extremely costly cases). The agency appears to be responding to a comment that
suggests that a shift to cost outliers would encourage system participants to manipulate the
system to increase their own payments—undermining a core purpose of shifting to the
Prospective Payment System in 1983 in the first place. In essence, in response, the agency
86
(1) acknowledges why the changes could allow participants to manipulate the system,
(2) explains why the agency believed that such manipulation was less likely than it appeared, and
(3) emphasized that there were incentives to manipulate both a day outlier and a cost outlier
system, such that manipulation per se was not a reason to refrain from shifting from a focus on
day outliers to a focus on cost outliers. The Court concludes that the agency sufficiently
responded to the issues presented and explained its decision to, as relevant here, adopt an
approach using hospital-specific cost-to-charge ratios based on the latest settled cost reports.
One final point is worth emphasizing with respect to this excerpt from the 1988
rulemaking. As the Court has reiterated before, judicial review of the 1988 regulation requires
using the judicial time machine and focusing on the record before the agency in 1988. Plaintiffs
interpret this comment as if the statement in the 1988 Federal Register about potential
manipulation referred to a small set of hospitals that could manipulate the system by raising
charges. However, that interpretation appears infected by hindsight. While the agency discovered
in 2003 that a small set of 123 hospitals had manipulated outlier payments by increasing
charges—hospitals that became known as “turbo-chargers,” see Dist. Hosp. Partners, 786 F.3d at
51—there is no indication that the agency, in 1988, was referring to a subset of bad actor
hospitals. Instead, it appears equally possible—if not likely—that the agency was referring to the
possibility that the new system would incentivize hospitals, as a whole, to increase their charges.
Indeed, that would be consistent with the problem that Congress attempted to solve just five
years beforehand in introducing the Prospective Payment System—the incentives in a cost-based
system to encourage hospitals generally to increase their costs in order to increase their
payments. See Cnty. of Los Angeles, 192 F.3d at 1008. Particularly in light of this possibility, the
agency rationally explained how it understood that increases in charges would be mitigated and
87
why, under this understanding, it chose to shift towards an approach focusing on cost outliers,
including the use of hospital-specific cost-to-charge ratios, rather than on day outliers.
Finally, in addition to pointing to comments by the agency that Plaintiffs’ claim are
inconsistent with the use of settled cost reports, Plaintiffs argue that the agency could have made
outlier payments subject to adjustment after cost reports for the applicable payment period were
finally settled—a process that the agency adopted in 2003 under the label reconciliation.
However, once again, Plaintiffs point to nothing in the record that shows that this possibility was
before the agency in 1988. Because Plaintiffs have not shown that this proposal was a
“significant and viable and obvious alternative[]” at the time of the 1988 rulemaking, this claim
fails. Nat’l Shooting Sports Found., Inc. v. Jones, 716 F.3d at 215.
In sum, the agency’s decision to use hospital-specific cost-to-charge ratios based on the
latest settled cost reports was neither arbitrary nor capricious. The agency rationally concluded
that this choice was the best among the “significant and viable and obvious alternatives” and
sufficiently explained its decision to do so.
Use of Statewide Average Cost-to-Charge Ratios
Plaintiffs argue that it was arbitrary and capricious to use statewide average cost-to-
charge ratios for hospitals where their individual cost-to-charge ratios were extremely low. As a
reminder, when the agency instituted the use of hospital-specific cost-to-charge ratios in 1988, it
concluded that “Statewide cost-to-charge ratios [would be] used in those instances in which a
hospital’s cost-to-charge ratio falls outside reasonable parameters.” 41 53 Fed. Reg. at 38,529
41
This provision was eliminated through the revisions to the outlier payment regulations
promulgated in 2003 for discharges occurring after August 7, 2003. See 68 Fed. Reg. at 34,500.
88
(codified at 42 C.F.R. § 412.84(h)). The agency would sets both those “parameters” and the
statewide cost-to-charge ratios annually. See id. (codified at 42 C.F.R. § 412.84(h)).
In the 1988 rulemaking, the agency determined that “[t]he range of reasonable cost-to-
charge ratios represents 3.0 standard deviations (plus or minus) from the mean of the log
distribution of cost-to-charge ratios for all hospitals.” Id. at 38,503. The agency also calculated
the upper and lower boundaries for cost-to-charge ratios. Id. The agency explained its reasoning:
We believe that ratios falling outside this range are unreasonable and are probably
due to faulty data reporting or entry. Therefore, they should not be used to identify
and pay cost outliers. We believe that 3.0 standard deviations represents an
appropriate range and that the accuracy of cost-to-charge ratios falling outside that
range is questionable.
53 Fed. Reg. at 38,507-08. This explanation was reasonable.
Once again, the fact that the agency later came to the conclusion that the use of statewide
cost-to-charge ratios allowed “turbo-chargers” to reap benefits from increasing charges does not
mean that the decision was arbitrary or capricious when made in 1988. Plaintiffs can point to
nothing that suggests that this decision was arbitrary and capricious when the regulation was
promulgated. Plaintiffs point only to a comment submitted in the context of the 1994 rulemaking,
claiming that the use of statewide ratios for hospitals with extremely low costs created an
incentive to inflate charges. See 59 Fed. Reg. at 45,407-08. While that concern ultimately
appears prescient, a comment raised in 1994 cannot be used, after the fact, to challenge the
regulation promulgated in 1988. In the 1994 rulemaking, the agency promulgated changes to the
outlier payment regulations to comply with the changes to the Medicare Act that Congress had
recently enacted. However, the agency never considered whether to eliminate the use of
statewide averages in that rulemaking. See 59 Fed. Reg. 27,708, 27,737-69 (May 27, 1994). As
explained before, just because the agency received and responded to comments outside of the
scope of the rulemaking does not expand the scope of that rulemaking. See Biggerstaff, 511 F.3d
89
at 186. In addition, insofar as the comment was presented in the context of setting the rates for
FY 1995, including the “reasonable parameters” outside of which statewide ratios would be used,
Plaintiffs have not challenged any payment determinations from FY 1995 in this action.
Moreover, Plaintiffs do not show any way in which decisions made specifically for FY 1995
infect subsequent fixed loss threshold rulemakings. Therefore, they are not relevant to Plaintiffs’
challenges to payment determinations in subsequent fiscal years. In any event, the agency
provided an adequate response to the comment on which Plaintiffs rely. The agency stated that it
did “not believe hospitals are setting their charges just to manipulate their cost-to-charge ratios.”
59 Fed. Reg. at 45,408. The agency further explained its reasoning, explaining that “contrary to
the commenter’s contention, the incentives a hospital would have to maximize outlier payments,
if any, would be to lower charges in order to increase its cost-to-charge ratio.” Id. While
Plaintiffs claim that the logic behind this explanation is faulty, the Court would conclude—even
if this comment were relevant to proceedings at issue in this case—that it was an adequate
response to the issue raised, particularly given the agency’s expertise in implementing the
complex statutory scheme with which the agency is entrusted. See Methodist Hosp. of
Sacramento v. Shalala, 38 F.3d 1225, 1229 (D.C. Cir. 1994) (“[I]n framing the scope of review,
the court takes special note of the tremendous complexity of the Medicare statute. That
complexity adds to the deference which is due to the Secretary’s decision.”).
Data Quality
Plaintiffs also argue that the agency did not implement the statewide cost-to-charge ratio
policy correctly. The Secretary responds that Plaintiffs incorrectly understand the record and
argue that the agency did not commit the alleged errors. The Court agrees with the Secretary.
90
First, Plaintiffs argue that the agency improperly used the statewide ratios from prior
years in calculating the fixed loss thresholds. Plaintiffs state that the Impact Files that the agency
used to calculate the fixed loss thresholds include the statewide ratio from a prior year for certain
hospitals. The Secretary acknowledges that the Impact Files include prior-year statewide
averages. However, she explains that, during the process of calculating the fixed loss threshold,
updated statewide averages were calculated and used in setting the fixed loss threshold. While
the administrative record does not include the computer programs used to perform the
calculations, the Court has no basis to doubt the Secretary’s representation. Moreover, the
agency adequately described the sources of the data for the model in the challenged rulemakings.
For example, the agency explained that to calculate FY 2003 fixed loss threshold, it “simulated
payments by applying FY 2003 rates and policies to the March 2002 update of the FY 2001
MedPAR file and the March 2002 update of the Provider-Specific File.” 67 Fed. Reg. at 50,122.
The agency further explained that, for hospitals where the hospital-specific cost-to-charge ratio
was more than 3.0 standard deviations above or below the mean of the log distribution of cost-to-
charge ratios for all hospitals, the agency used the statewide average cost-to-charge ratios found
in Tables 8A and 8B of the rulemaking. Id. at 50,125. In the FY 2002 rulemaking, as with the
other annual rulemakings, the agency used input from the several sources enumerated here to
calculate the fixed loss threshold and to calculate the statewide averages, all of which were
promulgated in that rulemaking. See id. at 50,122, 50,263 (promulgating Tables 8A and 8B for
FY 2002). This fact is consistent with the agency’s explanation in its briefing that the agency
calculated the statewide averages in the course of the modeling that generated the fixed loss
threshold and then used these revised averages in generating the fixed loss threshold. The agency
need not have done any more.
91
Second, Plaintiffs claim that, for certain hospitals, the agency used the statewide averages
themselves, instead of the hospital-specific cost-to-charge ratios, to compute the statewide
averages. The agency once again responds that this is incorrect. Plaintiffs do not point to any
support in the record for this claim. They point only to a document in the record showing those
facilities that defaulted to the statewide cost-to-charge ratios, A.R. (2003 amendments) 4417.332,
and to the list of statewide ratios for the FY 2003 rulemaking, A.R. (FY 2003) 4819.001. Once
again, the Court has no reason to disbelieve the agency’s representation that it did not use
statewide averages to calculate those averages themselves.
In sum, while the agency ultimately decided to change the features of the outlier payment
scheme that Plaintiffs challenge here—the use of settled cost reports and the use of statewide
averages for hospitals with low cost-to-charge ratios—the agency’s decision to implement these
features in the first instance was neither arbitrary nor capricious. It appears the agency
subsequently learned from its experience implementing these decisions, but that does not provide
cause for this Court to invalidate the agency’s 1988 action. Plaintiffs have pointed to nothing that
would show that the outlier payment regulations applicable to any of outlier payment
determinations challenged in this action were arbitrary or capricious.
2. Challenge to the FY 1998 through 2003 Fixed Loss Thresholds
Plaintiffs present two groups of challenges to the six fixed loss thresholds for FY 1998
through 2003. First, they claim that, for these fiscal years, the agency kept increasing the fixed
loss threshold despite evidence that the model “had no rational relationship to the real world.”
Pls.’ Mot. at 47. Second, Plaintiffs claim that the agency wrongly used “fudge factors” for FY
2001 through FY 2003. Neither arguments succeeds. Before addressing these two claims, the
Court notes that, notwithstanding Plaintiffs’ claim that they are challenging each fixed loss
92
threshold rulemaking from FY 1997 through FY 2003, they have clarified that they “do not claim
underpayments with respect to the FY 1997 [fixed loss threshold].” Pls.’ Mot. 48 n.38. Because
they are not claiming underpayments for that fiscal year, the Court does not understand Plaintiffs
to be arguing that the FY 1997 fixed loss threshold rulemaking was arbitrary or capricious. In
any event, Plaintiffs would have no basis to do so because they are not challenging any payment
determinations for that year.
a. Plaintiffs’ Tale of Continuing Increases in the Fixed Loss Threshold
In presenting their claim that the agency continued to increase the fixed loss threshold
between FY 1997 and FY 2003 despite evidence that the “model had no rational relationship to
the real world,” id., Plaintiffs have failed to heed the Court’s instruction—and the fundamental
principle of judicial review of agency action—that their challenges must be directed at discrete
agency actions and that they must be based on the administrative record before the agency at the
time of the respective actions. Plaintiffs tell a story of continually increasing fixed loss threshold
in the face of contradictory data, but do not show how any of these seven rulemakings were
arbitrary and capricious based on the record before the agency at the time of the rulemaking.
First, Plaintiffs emphasize how the agency kept missing its target for outlier payments
despite the increasing fixed loss thresholds. Plaintiffs are correct that, for several years, the
agency increased the fixed loss threshold, but nonetheless the outlier payments for that year
exceeded the 5.1 percent estimate. 42 However, this pattern does not show that the fixed loss
42
For example, for FY 1997, the agency set the fixed loss threshold at of $9,700—such that
outlier payments would be 5.1 percent of total DRG-based payments. See 61 Fed. Reg. at
46,228-29. As of July 31, 1998, the agency estimated that actual outlier payments for FY 1997
were 5.5 percent of total DRG-based payments. 63 Fed. Reg. at 41,009. For FY 1998, the agency
set the fixed loss threshold at $11,050—such that outlier payments would be 5.1 percent of total
DRG-based payments. 62 Fed. Reg. at 46,040. As of July 30, 1999, the agency estimated that
93
rulemakings were arbitrary and capricious. Despite Plaintiffs’ repeated emphasis that the agency
missed its targets, it was not required to “meet” those targets. As the Court of Appeals held in
County of Los Angeles, the agency is simply required to calculate the fixed loss thresholds such
that “when tested against historical data, will likely produce aggregate outlier payments totaling
between five and six percent of projected … DRG-related payments.” 192 F.3d at 1013. Just
because the agency was aware that actual outlier payments exceeded the predicted levels for
these years, see supra note 42, does not mean that it was arbitrary or capricious to continue
implementing this model. Plaintiffs present three arguments to the contrary; each is unavailing.
First, Plaintiffs claim that it should have been obvious to the agency that the increase in
outlier payments was caused by increases in charges. As the basis for this argument, Plaintiffs
rely on the following syllogism. The agency was aware that charges were rising faster than costs,
and thus that cost-to-charge ratios were (necessarily) declining. The agency had previously
explained that using a cost-inflation methodology automatically accounted for changes in the
cost-to-charge ratios “since the relevant variable is the costs estimated for a given case.” 59 Fed.
Reg. at 45,407. Therefore, Plaintiffs say, it should have been obvious that the actual increase in
outlier payments “was attributable to the principal remaining variable: hospital charges.” Pls.’
Mot. at 50-51. The Court disagrees. Under this complex statutory and regulatory scheme, the
Court would not conclude that the source of the increases was so obvious such that the agency
was under an obligation to address it sua sponte, without being prompted by a public comment.
Yet, despite Plaintiffs’ claim that the agency did not address the problem “[d]espite warnings,”
actual outlier payments for FY 1998 were 6.5 percent of total DRG-based payments. 64 Fed.
Reg. at 41,547. For FY 1999, the agency set the fixed loss threshold at $11,100—such that
outlier payments would be 5.1 percent of total DRG-based payments. 63 Fed. Reg. at 41,008. As
of August 1, 2000, the agency estimated that actual outlier payments for FY 1999 were 7.6
percent of total DRG-based payments. 65 Fed. Reg. at 47,114.
94
Plaintiffs do not point to any place in the record where such warnings were made with respect to
the relevant fiscal years. Pls.’ Mot. at 51.
Second, Plaintiffs claim that there was evidence that the model was producing irrational
results because several hospitals were projected to receive outlier payments that would be a high
percentage of the DRG-based payments. This argument fails for several reasons. Most
fundamentally, Plaintiffs cannot point to anywhere in the record where this phenomenon was
brought to the attention of the agency. Simply because the agency possessed data that could
reveal this phenomenon does not mean that the agency had an obligation to change its
methodology or to explain why it did not do so. In addition, Plaintiffs do not even attempt to
explain how they generated the calculations on which they rely for this argument. 43 Lastly, just
because agency set the fixed loss threshold such that the projected outlier payments for the entire
country would be 5.1 percent of the total projected nationwide DRG-based payments, pursuant to
the statutory requirements, does not mean it was arbitrary or capricious to set the threshold on
the basis of a prediction that would have generated high payments for individual hospitals. The
agency had no obligation to explain this phenomenon absent a relevant comment from a member
of the public or other evidence in the record showing that this issue was before the agency.
Third, Plaintiffs claim that the agency asserted, “in total contradiction of the record, that
there was no evidence of any problem.” Pls.’ Mot. at 52. However, it is Plaintiffs’ assertion that
is “in total contradiction of the record.” Plaintiffs’ sole citation to the record for this proposition
is a single response in the FY 2003 fixed loss threshold rulemaking. While Plaintiffs portray the
comment and response as pertaining to general delays in updates to cost-to-charge ratios, it
43
Plaintiffs cite to an additional exhibit, Exhibit 8, which the Court concluded above could not
be considered. In any event, Plaintiffs neither explain in their exhibit nor in the body of their
brief how they derived these calculations.
95
appears from the record that the exchange was limited to specific delays with the processing of
the FY 2000 cost reports. See 67 Fed. Reg. at 50,124. In that light, the agency responded that it
did not expect these additional delays—beyond the usual lag in the data processing—to have
detrimental effects. 44 See id. at 50,124-25. This response is adequate with respect to the problem
presented in the comment and does not show a general obliviousness to problems with the data
as Plaintiffs suggest.
In sum, Plaintiffs have failed to show that any of the fixed loss threshold rulemakings
from FY 1997 to FY 2003 were arbitrary or capricious because of the purported trend of
escalating fixed loss thresholds combined with outlier payments that continued to exceed the
agency’s projections.
b. Use of “Fudge Factors”
Plaintiffs claim that the agency used unexplained “fudge factors” in setting the fixed loss
thresholds for FY 2001 through FY 2003. The Court disagrees.
Plaintiffs first claim that the agency used an unexplained “fudge factor” in setting the FY
2001 fixed loss threshold. This is simply at odds with the record. As the agency recounted in
finalizing the FY 2001 fixed loss threshold, the agency had used a cost inflation factor of
negative 1.724 percent in projecting the costs for FY 1999, and the agency had used a zero
inflation factor in projecting the costs for FY 2000. See 65 Fed. Reg. at 47,113. For FY 2001, the
agency had proposed an inflation factor of 1.0 percent, but ultimately decided that a factor of 1.8
44
The following is the agency’s full response: “Our analysis shows that, despite the delay in
processing cost reports alluded to above, the average cost-to-charge ratios have continued to
decline. We note there is always a lag between the timeframe from which the cost-to-charge
ratios are taken and the period to which they are applied to charges. We do not have any evidence
that the higher than expected outlier payments result from any extra lag in updating cost-to-
charge ratios due to the delay in processing the cost reports.” 67 Fed. Reg. at 50,124-25.
96
percent was proper. See id. The agency explained the reasoning for using this inflation factor:
“This factor reflects our analysis of the best available cost report data as well as calculations
(using the best available data) indicating that the percentage of actual outlier payments for FY
1999 is higher than we projected before the beginning of FY 1999, and that the percentage of
actual outlier payments for FY 2000 will likely be higher than we projected before the beginning
of FY 2000.” Id. In other words, the agency explained why it used the higher inflation factor in
light of its experience in previous years. This is neither arbitrary nor capricious. The agency
explained why it used this factor, and it need not explain in any further detail exactly how its
analysis of the underlying data generated the 1.8 percent figure. See Tex. Mun. Power 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.”).
Plaintiffs present a similar argument regarding the FY 2002 threshold, and it fails for a
similar reason. 45 In setting the FY 2002 threshold, the agency recounted the cost inflation factors
used for the previous several years. See 66 Fed. Reg. at 39,941. While the agency had proposed
using a cost inflation factor for FY 2002 of 5.5 percent, the agency ultimately concluded that a
cost inflation factor of 2.8 percent was proper. See id. Once again the agency explained its
conclusion in light of its recent experience: “This factor reflects our analysis of the best available
cost report data as well as calculations (using the best available data) indicating that the
percentage of actual outlier payments for FY 2000 is higher than we projected before the
45
The Court notes that Plaintiffs do not cite any record evidence for this claim.
97
beginning of FY 2000, and that the percentage of actual outlier payments for FY 2001 will likely
be higher than we projected before the beginning of FY 2001.” Id. As with the FY 2001 cost
inflation factor, the agency adequately explained the choices behind its decision to choose this
factor. No more is necessary to survive arbitrary and capricious review. See Tex. Mun. Power, 89
F.3d at 869-70.
Finally, with respect to this set of years, Plaintiffs argue that the decision in the FY 2003
fixed loss threshold rulemaking to switch from cost inflation to charge inflation was arbitrary and
capricious. Plaintiffs primarily rely on the agency’s prior decision, in 1994, to switch from
charge inflation to cost inflation. Neither choice was arbitrary or capricious. To explain why, it is
necessary to explain briefly the differences in the two methodologies. Under the charge inflation
methodology, which the agency introduced for FY 2003, the agency calculated a measure of past
charge inflation based on historical data and used this measure to inflate past charges in order to
generate a dataset of projected charges for the fiscal year in question; the agency then adjusted
these charges to projected future costs using cost-to-charge ratios. See 70 Fed. Reg. at 47,495. By
contrast, under the cost inflation methodology, which was used for FY 1994 through FY 2002,
the agency adjusted past charges by cost-to-charge ratios to estimate past costs, and then used a
cost inflation factor derived from historical data to inflate the estimated costs and generate
projected future costs. See 58 Fed. Reg. at 46,347. Under either methodology, the agency
ultimately used the projected future costs to simulate outlier payments for the fiscal year in
question.
In 1994, the agency determined that it was best to switch from a charge inflation
methodology to a cost inflation methodology. 58 Fed. Reg. at 46,347. The agency reasoned:
However, we have noted a continued trend with respect to changes in costs
relative to charges. Over time, charges have continued to increase at a faster rate
98
than costs, so that cost-to-charge ratios have been declining. Because we use the
latest available cost-to-charge ratios (which may be as much as 2 years old) to
convert billed charges to costs for purposes of estimating cost outlier payments,
we may be overestimating outlier payments in setting the thresholds. As a result,
actual payments may be lower than estimated. In order to alleviate this problem,
we are using a cost inflation factor rather than a charge inflation factor to estimate
FY 1994 costs. In other words, instead of inflating the FY 1992 charge data by a
charge inflation factor for 2 years in order to estimate FY 1994 charge data and
then applying the cost-to-charge ratio, we will adjust the charges by the cost-to-
charge ratio and then inflate the estimated costs for 2 years of cost inflation. In
this manner, we will be automatically adjusting for any changes in the cost-to-
charge ratios that may occur, since the relevant variable is the costs estimated for
a given case.
Id. Plaintiffs do not challenge this choice. Almost ten years later, the agency determined that, in
light of the agency’s experience, it would be preferable to make the opposite choice and use a
charge inflation methodology. See 67 Fed. Reg. at 50,124. The agency explained its decision:
Based on our analysis above, we believe that, due to current trends in hospital
charging practices, using inflation factors based on annual cost growth results in
underestimating the percentage of outlier payments. That is, if charges are
growing at a faster rate than costs, inflating FY 2001 charges by the observed rate
of change in costs will underestimate FY 2003 charges, thereby resulting in
outlier payments greater than 5.1 percent. … Because charge data are available
from claims data in the MedPAR file, they are more up-to-date than cost data
taken from the cost reports.
Id. In short, even though the agency was trying to resolve challenges in accounting for cost and
charge inflation at both junctures, neither approach was arbitrary or capricious. The agency
adequately explained its decision in both circumstances, and the agency’s hands were not tied in
2002 simply because it had previously decided to switch from charge inflation to cost inflation.
See Fox Television Studios, 556 U.S. at 515 (“But [an agency] need not demonstrate to a court’s
satisfaction that the reasons for the new policy are better than the reasons for the old one; it
suffices that the new policy is permissible under the statute, that there are good reasons for it, and
that the agency believes it to be better, which the conscious change of course adequately
99
indicates.”). Accordingly, the agency’s decision to adopt a charge inflation methodology in 2003
was neither arbitrary nor capricious.
In sum, none of Plaintiffs’ challenges to the fixed loss threshold rulemakings from FY
1997 through FY 2003 are successful. Plaintiffs have not shown, based on the record, that any of
those rulemakings were arbitrary or capricious.
3. Challenges to the Fixed Loss Thresholds after the 2003 Changes to the Outlier
Payment Regulations
As described above, in 2003, the agency promulgated several changes to the outlier
payment regulations to address problems that had been discovered regarding the outlier payment
program. In short, the agency introduced three changes: (1) more recent data would be used to
derive the cost-to-charge ratios (specifically, tentatively settled cost reports would be used when
they were more recent than settled cost reports), (2) the agency would no longer use statewide
cost-to-charge ratios for hospitals with extremely low cost-to-charge ratios, and (3) outlier
payments would become subject to adjustment, in an after-the-fact reconciliation process, when
the actual cost-to-charge ratios deviated substantially from the ratios used to make the outlier
payments. See 68 Fed. Reg. at 34,497-503. These changes were in effect for FY 2004 and
beyond. 46 See id. at 34,498, 34,510; see also 42 C.F.R. § 412.84(h), (i). Plaintiffs challenge the
agency’s decision not to implement a mid-year adjustment to the FY 2003 fixed loss threshold
46
More precisely, the use of statewide cost-to-charge ratios for hospitals with extremely low
cost-to-charge ratios was eliminated for discharges occurring on or after August 8, 2003, less
than two months before the beginning of FY 2004. 68 Fed. Reg. at 34,510. For discharges
occurring on or after October 1, 2003, the latest of the settled or the tentatively settled cost
reports would be used to derive the cost-to-charge ratios. See id. at 34,498. For discharges
between August 8, 2003, and October 1, 2003, outlier payments were only partially subject to
reconciliation in light of the fact that the use of tentatively settled cost reports was not yet in
effect. See 42 C.F.R. § 412.525(a)(4); 68 Fed. Reg. at 34,515. For discharges on or after October
1, 2003, outlier payments would be fully subject to reconciliation. See id.
100
along with these changes to the outlier payment regulations. Plaintiffs also challenge each of the
fixed loss threshold rulemakings for the fiscal years for which these new rules were in effect: FY
2004 through FY 2007.
Plaintiffs raise three general arguments with respect to these four rulemakings. First, they
claim that, because the charge data on which these rulemakings were based was from the “turbo-
charging” era, the agency should have adjusted the data in setting the fixed loss thresholds. Pls.’
Mot. at 55. Second, Plaintiffs argue that, even with the use of more recent data to derive the cost-
to-charge ratios (from tentatively settled cost reports), because of the lag between the setting of
the fixed loss threshold and the calculation of actual outlier payments, the agency should have
adjusted the cost-to-charge ratios to account for continuing declines in those ratios. Id. at 55-56.
Third, Plaintiffs argue that the fixed loss threshold for each of these years was flawed because
the agency failed to account for reconciliation. Id. at 56. Because each of these arguments applies
differently to the four fixed loss threshold rulemakings during this period, it is necessary to
evaluate Plaintiffs’ specific arguments directed at each of the annual rulemakings in light of the
data before the agency at the relevant times. Before doing so, it is necessary to resolve one
preliminary issue. Several of Plaintiffs’ challenges are based on the draft interim final rule that
was sent to the Office of Management and Budget in 2003. See id. at 56-57. However, the Court
of Appeals confirmed that the draft rule cannot be used to challenge any agency rulemakings
because it was never published in the Federal Register. See Dist. Hosp. Partners, 786 F.3d at 58.
Accordingly, insofar as Plaintiffs’ challenges are based on the draft interim final rule, they
cannot succeed. With this in mind, the Court first addresses Plaintiffs’ challenge to the agency’s
decision not to undertake a mid-year adjustment to the FY 2003 fixed loss threshold and then
101
addresses, in turn, Plaintiffs’ specific challenges to the fixed loss threshold rulemakings for FY
2004 through FY 2007.
a. Mid-year Adjustment to the FY 2003 Fixed Loss Threshold
In the agency’s 2003 rulemaking promulgating changes to the outlier payment
regulations, the agency responded to comments suggesting that the agency make a mid-year
adjustment to the FY 2003 fixed loss threshold and explained why it decided not to make such an
adjustment. 68 Fed. Reg. at 34,505. Plaintiffs now argue that it was arbitrary and capricious for
the agency to decline to make such a mid-year adjustment.
Plaintiffs’ arguments are largely based on the draft interim final rule that the agency
transmitted to the Office of Management and Budget in February 2003. See AR (2003
amendments) 4417.338. However, the agency abandoned the plan to promulgate an interim final
rule and never published such a rule in the Federal Register. Accordingly, the agency had no
obligation to explain its departure from proposals in that draft rule, such as the proposal to
implement a mid-year adjustment to the FY 2003 threshold. See Dist. Hosp. Partners, 786 F.3d
at 58 (“It is true, of course, that an agency cannot ‘depart from a prior policy sub silentio or
simply disregard rules that are still on the books.’ But this principle is inapplicable here—the
OMB draft was never ‘on the books’ in the first place.”) (quoting Fox Television Stations, Inc.,
556 U.S. at 515).
Nonetheless, in response to public comments the agency received, the agency considered
the appropriateness of a mid-year adjustment. To determine whether an adjustment was
appropriate, the agency re-estimated the fixed loss threshold for FY 2003 using the more recent
data then available—FY 2002 data rather than data from earlier fiscal years—and taking into
account the fact that some of the changes to the outlier payment regulations promulgated in that
102
same rule would be in effect during the last two months of FY 2003. 68 Fed. Reg. at 34,505. The
agency estimated the “threshold would be only slightly higher than the current threshold
[$33,560] (by approximately $600).” Id.
Plaintiffs argue that the analysis that generated this conclusion was arbitrary and
capricious, but their arguments fail. They argue that the agency improperly applied the “old”
fixed loss threshold model, by which they appear to mean the model that was used to calculate
the fixed loss thresholds for the fiscal years prior to the 2003 revisions to outlier payment
regulations. However, none of the purported flaws that they identify show that the agency’s
choice of model was arbitrary or capricious. First, Plaintiffs claim that the model provided
“statutorily unauthorized payments” to turbo-charging hospitals. However, as the Court
explained above with respect to Plaintiffs’ challenges invoking Chevron, payments to turbo-
chargers were not in fact statutorily prohibited. Second, Plaintiffs claim that the agency wrongly
included the data pertaining to the turbo-chargers—the hospitals that rapidly increased their
charges—in computing its inflation factor. Third, Plaintiffs argue that the agency mistakenly
failed to account for continuing declines in the cost-to-charge ratios when it decided not to adjust
those ratios. However, with respect to Plaintiffs’ second and third arguments, Plaintiffs do not
point to anywhere in the administrative record for the 2003 rulemaking where these issues were
raised before the agency. While the agency discussed several comments regarding a mid-year
adjustment to the FY 2003 threshold, none of the comments raised the specific issues that
Plaintiffs now raise. Accordingly, the agency had no obligation to consider these specific issues
in re-estimating the FY 2003 fixed loss threshold. Plaintiffs also highlight the agency’s statement
that it “inflated charges from the FY 2002 Medicare Provider Analysis and Review (MedPAR)
file by the 2-year average annual rate of change in charges per case to predict charges for FY
103
2004.” 68 Fed. Reg. at 34,505 (emphasis added). As the agency explains, this reference to FY
2004 is self-evidently a typographical error. Every other reference in the discussion was to the
proper year, including its statement that the agency concluded “it would be appropriate to use FY
2002 data to reestimate the FY 2003 threshold.” Id. There is no reason to believe that, contrary to
the agency’s representation in its briefing, that the agency inflated data in order to project cost
for FY 2004 rather than for FY 2003.
Aside from Plaintiffs’ specific claims regarding the purported flaws with the agency’s re-
calculation of the FY 2003 fixed loss threshold, Plaintiffs ignore the fact that the 2003 revisions
to the outlier payment regulations were generally not applicable for FY 2003. See supra note 46.
The only change that became fully applicable during even a part of that fiscal year was the
elimination of the use of statewide cost-to-charge ratios for hospitals with extremely low cost-to-
charge ratios—and only for discharges occurring on or after August 8, 2003. 47 68 Fed. Reg. at
34,510. Even this change was applicable for less than two months of FY 2003. Accordingly,
notwithstanding Plaintiffs’ suggestion to the contrary, see Pls.’ Mot. at 58 n.44, one would not
necessarily expect that applying “the changes implemented in this final rule that will be in effect
during a portion of FY 2003” would generate a significant change in the fixed loss threshold for
FY 2003.
In light of its conclusion that a re-estimated fixed loss threshold would be slightly higher
than the existing threshold for FY 2003, the agency explained why it concluded that a mid-year
adjustment was not warranted:
We believe it is appropriate not to change the FY 2003 outlier threshold at this
time. Although our current empirical estimate of the threshold indicates it could
be slightly higher, there are other considerations that lead us to conclude the
47
The outlier payments for the period from August 8, 2003, to September 30, 2003, were only
partially subject to reconciliation. See 68 Fed. Reg. at 34,498, 34,510.
104
threshold should remain at $33,560. Increasing the threshold would result in
lower outlier payments for all hospitals, not just those that have been aggressively
maximizing their outlier payments. Changing the threshold for the remaining few
months of the fiscal year could disrupt hospitals’ budgeting plans and would be
contrary to the overall prospectivity of the [Prospective Payment System]. We do
believe that we have the authority to revise the threshold, given the extraordinary
circumstances that have occurred (in particular, the manipulation of the policy by
some hospitals). However, in light of the relatively small difference between the
current threshold and our revised estimate, and the limited amount of time
remaining in the fiscal year, we have concluded it is more appropriate to maintain
the threshold at $33,560.
Id. at 34,506. This explanation is adequate on its own terms, and there is no need to consider it
further. Indeed, Plaintiffs do not challenge the agency’s decision not to raise the fixed loss
threshold; they only argue that, if the fixed loss threshold had been “properly” re-calculated, it
should have been lowered. Accordingly, the Court concludes that the agency’s decision not to
implement a mid-year adjustment to the FY 2003 fixed loss threshold was not arbitrary or
capricious. 48
b. Challenge to the FY 2004 Fixed Loss Threshold Rulemaking
With respect to the FY 2004—the first full fiscal year where the 2003 revisions to the
outlier payment regulations were applicable—Plaintiffs argue that the decision to set the fixed
loss threshold at $31,000 was arbitrary and capricious (1) because the data used to set the fixed
loss threshold included data infected by turbo-charging, (2) because the agency failed to adjust
the cost-to-charge ratios to account for continuing declines in cost-to-charge ratios, and
(3) because the agency failed to account for reconciliation. See Pls.’ Mot at 62.
48
Because the Court concludes that the agency adequately justified its decision not to implement
a mid-year adjustment to the FY 2003 fixed loss threshold, it need not reach Defendant’s
alternative argument that the decision whether to make such an adjustment is committed to the
agency’s discretion by law.
105
Before addressing the individual arguments, the Court notes that, once again, Plaintiffs
rely substantially on the draft interim final rule to impugn the FY 2004 fixed loss threshold
rulemaking. However, they may not do so because that rule was never published in the Federal
Register. See Dist. Hosp. Partners, 786 F.3d at 58. Similarly, Plaintiffs may not rely on
testimony given by Thomas Scully, Administrator of the Center for Medicare and Medicaid
Services, before a Congressional subcommittee to impugn subsequent rulemakings. Plaintiffs
cite a statement of Scully in testimony before the Senate Appropriations Subcommittee on Labor,
Health and Human Services, and Education, and Related Agencies, on March 11, 2003, that “the
correct number [for the fixed loss threshold] probably is in the midtwenties.” See Medicare
Outlier Payments to Hospitals: Hearing Before Subcomms. On Appropriations and Labor,
Health & Human Services, and Education, 108th Cong. 108-268, at 13 (2003) (statement of
Thomas A. Scully, Adm’r, Centers for Medicare & Medicaid Servs., Dep’t of Health & Human
Servs.), available at http://www.gpo.gov/fdsys/pkg/CHRG-108shrg85832/pdf/CHRG-
108shrg85832.pdf (last visited on Aug. 3, 2015). It appears that this testimony represents
Scully’s personal opinion rather than the official position of the agency. See id. (“I happen to
believe, and our actuaries believe that the correct number probably is in the midtwenties, if we
fix the program abuses, I really think that fixing this will provide the other 97 percent of the
hospitals that have not abused more money from it. And so I do think the outlier threshold—my
personal opinion is that it probably, if we fix the abuses, would be too high, but I can understand
the skepticism from our budget analysts to say we have been wrong 5 years in a row by a couple
of billion dollars, how could we possibly think we are right now?”) (emphasis added). In any
event, Scully’s speculation about the appropriate level for the fixed loss threshold cannot be used
to attack the FY 2004 fixed loss threshold that was ultimately promulgated, particularly when he
106
testified before the promulgation of the final 2003 revisions to the outlier payment regulations—
which governed FY 2004 and beyond—and before the agency conducted the analysis that
resulted in the fixed loss threshold promulgated for FY 2004. Furthermore, the agency has no
obligation to explain why the fixed loss threshold for this fiscal year deviated from the
administrator’s previous speculations. Cf. Dist. Hosp. Partners, 786 F.3d at 58 (no obligation to
explain deviation from policy that was never “on the books”). With that in mind, the Court
addresses the substance of Plaintiffs’ three challenges regarding the FY 2004 fixed loss
threshold.
First, Plaintiffs argue that the agency should have excluded the data that was related to
turbo-chargers from the charge inflation calculation. In addition to their reliance on the findings
of the draft interim final rule, Plaintiffs argue that the agency never responded to a comment
presenting such a suggestion. Indeed, the record reflects that one commenter “requested that
CMS factor in the calculation of the threshold the fact that certain hospitals have distorted their
charges significantly.” 68 Fed. Reg. at 45,477. Plaintiffs argue that the agency did not adequately
respond to this comment and did not explain why it chose to include data from the turbo-chargers
when calculating the charge inflation factor for FY 2004. In District Hospital Partners, the D.C.
Circuit Court of Appeals concluded that the agency had not adequately explained why it did not
exclude the 123 turbo-charging hospitals from the charge inflation calculation for that fiscal year.
See 786 F.3d at 58, 59. The Court of Appeals concluded that the FY 2004 rule must be remanded
to the agency for further explanation regarding the inclusion of the 123 turbo-charging hospitals
that had been referenced in the notice of proposed rulemaking issued by the agency earlier in
2003. See id. at 60. This Court is, therefore, constrained to do the same and will remand the FY
2004 rule to the agency to provide the agency an opportunity to explain further why it did not
107
exclude the 123 identified turbo-charging hospitals from the charge inflation calculation for FY
2004—or to recalculate the fixed loss threshold if necessary. While the Court must remand the
FY 2004 rulemaking to the agency because of the gap in its contemporaneous explanation in that
rulemaking, the Court nonetheless considers Plaintiffs’ other claims regarding the FY 2004
rulemaking to determine the scope of that remand.
Second, Plaintiffs claim that the agency improperly failed to include an adjustment factor
in the cost-to-charge ratio despite the fact that cost-to-charge ratios continued to decline. Aside
from Plaintiffs’ impermissible reliance on the interim final rule, they support this argument by
pointing to a single comment submitted by the Federation of American Hospitals on July 8,
2003. Above, the Court concluded that it would not be proper, at this point, to supplement the
administrative record with this comment because of its untimely submission. Without that
comment, Plaintiffs’ cannot point to any evidence in the record showing that the use of an
adjustment factor with respect to the cost-to-charge ratios was a “‘significant and viable and
obvious alternative[]’” at the time of the agency’s decision. Dist. Hosp. Partners, 786 F.3d at 59
(citation omitted). For that reason alone, the Court concludes that this particular challenge fails.
In any event, even if the Court considered the Federation of American Hospitals
comment, which the Court has excluded from the record, the Court would conclude that the
agency’s decision not to apply an adjustment factor to the cost-to-charge ratios across the board
was neither arbitrary nor capricious. See Dist. Hosp. Partners, 973 F. Supp. 2d at 15-16. Notably,
the agency did, in fact, attempt to account for declining cost-to-charge ratios with respect to a
certain subset of hospitals “that ha[d] been consistently overpaid recently for outliers.” 68 Fed.
Reg. at 45,476. For those hospitals, the agency “attempted to project each hospital’s cost-to-
charge ratio based on its rate of increase in charges per case based on FY 2002 charges,
108
compared to costs (inflated to FY 2002 using actual market basket increases).” Id. at 45,477. For
other hospitals, the agency used the cost data from the most recent cost-reporting year to
approximate the cost-to-charge ratios that would be used for actual outlier payments in FY 2004,
which would be based on the latest tentatively settled cost reports. See id. at 45,476. It was
rational for the agency to limit its attempt to project future decreases in cost-to-charge ratios to a
small subset of hospitals where recent history strongly suggested that the ratios for those
hospitals would continue to decrease—but to continue to use actual data in deriving the cost-to-
charge ratios for all other hospitals. In light of the Court’s deferential standard of review, see
West Virginia, 362 F.3d at 868, even if the Court were to consider 2004 Federation of American
Hospitals comment as part of the administrative record, the Court would not invalidate the
agency’s decision to use actual historical data and to limit its projections of future data changes
to a small subset of Medicare facilities.
Third, Plaintiffs argue that the agency improperly failed to consider the effects of
reconciliation on the projections for FY 2004. Once again, in support of this argument, Plaintiffs
rely only on a statement in the Federation of American Hospitals comment—which is not part of
the administrative record—and on the draft interim final rule. As explained above, neither can
serve as the basis for Plaintiffs’ claim that the agency’s action was arbitrary and capricious for
failing to account for reconciliation. Nonetheless, the Court notes that this challenge would also
fail because the agency did account for the effects of reconciliation for FY 2004. See 68 Fed.
Reg. at 45,476-77. The agency noted that it “is difficult to project which hospitals will be subject
to reconciliation of their outlier payments using available data.” Id. at 45,476. Nonetheless, based
on the data available, the agency explained how it had accounted for the effects of reconciliation:
Based on our analysis of hospitals that have been consistently overpaid recently
for outliers, we have identified approximately 50 hospitals we believe will be
109
reconciled. Therefore, for these hospitals, to account for the fact that the
reconciliation will result in different outlier payments than predicted using the
cost-to-charge ratios calculated as described above, we attempted to project each
hospital’s cost-to-charge ratio based on its rate of increase in charges per case
based on FY 2002 charges, compared to costs (inflated to FY 2002 using actual
market basket increases).
Id. at 45,476-77. The agency also noted that “the amount of fiscal intermediary resources
necessary to undertake reconciliation will ultimately influence the number of hospitals
reconciled.” Id. at 45,476. Taken together, these statements adequately explain why the agency
accounted for reconciliation with respect to the set of 50 hospitals rather than with respect to all
123 hospitals previously identified as turbo-chargers.
For all of these reasons, the Court remands the FY 2004 fixed loss threshold rulemaking
to the agency to allow it to explain further why the agency did not exclude the 123 turbo-
charging hospitals from its calculation of charge inflation. See Dist. Hosp. Partners, 786 F.3d at
60 (concluding that remand is appropriate to allow agency to articulate better explanation of
decision). The Court rejects all of Plaintiffs’ other challenges to the FY 2004 fixed loss threshold.
c. Challenge to the FY 2005 Fixed Loss Threshold Rulemaking
With respect to the FY 2005 fixed loss threshold rulemaking, Plaintiffs raise once again
the challenges that they raise with respect to the FY 2004 fixed loss threshold rulemaking—that
the agency used data infected by turbo-charging when the setting the fixed loss threshold, that
the agency should have used an adjustment factor to account for declining cost-to-charge ratios,
and that the agency failed to account for reconciliation. In addition, Plaintiffs argue that the
agency should have used a cost inflation methodology rather than a charge inflation
methodology. The Court addresses these arguments in turn.
110
First, Plaintiffs argue that the agency failed to address the trend of declining cost-to-
charge ratios. However, in responding to a comment suggesting the use of such an adjustment
factor, the agency explained why it concluded it was not necessary to use such a factor:
We do not believe that it is necessary to make a specific adjustment to our
methodology for computing the outlier threshold to account for any decline in
cost-to-charge ratios in FY 2005, as the commenter has requested. We have
already taken into account the most significant factor in the decline in cost-to-
charge ratios, specifically, the change from using the most recent final settled cost
report to the most recent tentatively settled cost report. Furthermore, we strongly
prefer to employ actual data rather than projections in estimating the outlier
threshold because we employ actual data in updating charges, themselves.
69 Fed. Reg. at 49,277. While Plaintiffs critique this explanation, the Court concludes it is
sufficient. It is not for the Court to second guess the agency’s choices when it has, as here,
provided a cogent explanation for those choices. Indeed, while Plaintiffs suggest that it was
irrational for the agency to prefer to use actual cost-to-charge ratios over projected ratios when it
was already using projected data, the Court disagrees. It is reasonable for the agency to prefer to
use only projected data for certain factors, as it has explained here. In addition, the agency
explained that it had addressed the most significant factor regarding declining cost-to-charge
ratios by using more recent data. Altogether the agency sufficiently explained its conclusion that
an adjustment factor was not necessary.
Second, Plaintiffs argue that the charge inflation factor used was flawed. Plaintiffs argue
that the charge inflation factor was distorted by data from turbo-charging hospitals and that the
agency did not adequately address its choice to continue using a charge inflation methodology
rather than reverting to a cost inflation methodology.
To assess these arguments, a brief review of the agency’s methodology to establish the
fixed loss threshold for FY 2005 is necessary. The agency used the FY 2003 charge data as a
baseline. 69 Fed. Reg. at 49,277. To develop a charge inflation factor, the agency took the
111
“unprecedented step of using the first half-year of data from FY 2003 and comparing data to the
first half year of FY 2004.” Id. Using this data, the agency calculated a one-year annual rate of
charge inflation of 8.9772, or 18.76 percent over two years, and then inflated the FY 2003 data
by this two-year charge inflation figure. Id. The agency then converted these projected charges
into projected costs by using hospital-specific cost-to-charge ratios from the April 2004 update to
the Provider Specific File. Id.
The D.C. Circuit Court of Appeals already considered and rejected the argument
presented by Plaintiffs here: that it was arbitrary and capricious to fail to exclude the turbo-
charging data from the calculation of a charge inflation factor for FY 2005. See Dist. Hosp.
Partners, 786 F.3d at 61-62. In District Hospital Partners, the Court of Appeals noted that, while
the data from the first half of FY 2003 was affected by turbo-charging, the charge data from the
first half of FY 2004 was not similarly affected because the underlying discharges had occurred
after the implementation of the 2003 changes to the outlier payment regulations. Id. at 61. The
Court of Appeals approved the agency’s decision to retain the turbo-chargers in the dataset in
order to ensure that the FY 2003 and FY 2004 datasets were comparable. See id. So, too, here. As
stated by the Court of Appeals in District Hospital Partners, it was not arbitrary or capricious to
include the turbo-chargers in the datasets used to calculate the charge inflation factor for FY
2005.
Plaintiffs also argue that it was arbitrary and capricious for the agency to reject a
suggestion that the agency ought to revert to a cost inflation methodology, which the agency had
used through FY 2002. In explaining its choice to continue using a charge inflation methodology,
the agency explained that “the use of charges is still appropriate because the basic tendency of
charges to increase faster than costs is still evident.” 69 Fed. Reg. at 49,277. Plaintiffs argue that
112
this explanation is insufficient because it was similar to the explanation gave in 1993 when the
agency shifted from a charge inflation methodology to a cost inflation methodology. However, as
the Court explained above, this explanation is adequate. Neither choice of methodology was
arbitrary or capricious, and the agency is not barred from reversing its earlier decision as long as
it has explained why it is making the present decision—as it has done here. See Fox Television
Studios, 556 U.S. at 515.
Third, Plaintiffs argue that the FY 2005 fixed loss threshold rulemaking was arbitrary and
capricious because the agency failed to account for reconciliation. However, the agency
explained at length why it was not accounting for reconciliation:
We are not including in the calculation of the outlier threshold the possibility that
hospitals’ cost-to-charge ratios and outlier payments may be reconciled upon cost
report settlement. Reconciliation occurs when hospitals’ cost-to-charge-ratios at
the time of cost report settlement are different than the tentatively settled cost-to-
charge-ratio used to make outlier payments during the fiscal year. However, we
believe that due to changes in hospital charging practices following
implementation of the new outlier regulations in the June 9, 2003 final rule, the
majority of hospitals’ cost-to-charge ratios will not fluctuate significantly enough
between the tentatively settled cost report and the final settled cost report to meet
the criteria to trigger reconciliation of their outlier payments. Furthermore, it is
difficult to predict which specific hospitals may be subject to reconciliation in any
given year. As a result, we believe it is appropriate to omit reconciliation from the
outlier threshold calculation.
69 Fed. Reg. at 49,278. This explanation is more than adequate. Plaintiffs emphasize that
reconciliation never occurred for FY 2005. However, that does not undermine the agency’s
explanation, which emphasizes uncertainty regarding reconciliation and reflects the agency’s
belief that the other changes to the outlier payment regulations obviated much of the need for
reconciliation. The agency has explained why it chose not to account for reconciliation given the
data that was available and given uncertainty about the implementation of the reconciliation
provision.
113
For all of these reasons, the Court concludes that none of Plaintiffs’ challenges to the FY
fixed loss threshold rulemaking succeed and that the FY 2005 fixed loss threshold rulemaking
was neither arbitrary nor capricious. 49
d. Challenge to the FY 2006 Fixed Loss Threshold Rulemaking
Next, the Court turns to Plaintiffs’ challenges to the FY 2006 fixed loss threshold
rulemaking. Before addressing Plaintiffs’ individual challenges, the Court notes that the D.C.
Circuit Court of Appeals concluded in District Hospital Partners that “the 2006 outlier threshold
was plainly reasonable.” 786 F.3d at 62. In addition, just like the Court of Appeals, this Court is
“perplexed” by Plaintiffs’ objection to this fixed loss threshold. See id. at 63. Plaintiffs “cite[]
favorably a comment submitted during the 2006 rulemaking that advocated a fixed loss threshold
of $24,050.” Id.; see Pls.’ Mot. at 66 (citing A.R. (FY 2006) at 658 (comment of Federation of
American Hospitals)). In fact, agency set the threshold lower than this proposed level—at
$23,600—which would therefore would generate outlier payments higher than would have
resulted from the proposal by the Federation of American Hospitals. See 70 Fed. Reg. at 47,494.
Nonetheless, the Court addresses the three arguments Plaintiffs present, which largely
recapitulate the arguments they lodged with respect to the FY 2005 fixed loss threshold.
Specifically, Plaintiffs argue (1) that the agency did not adequately explain its use of a charge
inflation methodology, (2) that the agency did not adequately account for the decline of the cost-
to-charge ratios over time, and (3) that the agency did not adequately account for reconciliation.
49
Plaintiffs also emphasize the fact that it appeared that the outlier payments for FY 2004 would,
in fact, be less than 5.1 percent of total DRG-based payments. However, the Court notes that the
mere fact that it appeared that FY 2004 outlier payments would be less than 5.1 percent of the
DRG-based payments does not make the agency’s failure to adjust its methodology arbitrary or
capricious. Given that the Court rejects Plaintiffs’ other challenges to the FY 2005 rulemaking, it
is of no moment that the data available at the time of the FY 2005 rulemaking suggested that FY
2004 outlier payments would be lower than the previous projections suggested.
114
Pls.’ Mot. at 66-68. As with Plaintiffs’ challenge to the FY 2005 rulemaking, none of these
arguments are persuasive.
First, Plaintiffs argue that the agency did not adequately respond to comments suggesting
a return to the use of a cost inflation methodology and did not adequately explain the continued
use of a charge inflation methodology. The Court disagrees. The agency responded to those
comments and explained that it “continue[d] to believe that using charge inflation, rather than
cost inflation, will more likely result in an outlier threshold that leads to outlier payments
equaling 5.1 percent of total [Prospective Payment System] payments.” 70 Fed. Reg. at 47,495.
The agency further explained why it was appropriate to use a charge inflation methodology in
establishing the fixed loss threshold. See id. Notably, in their brief, Plaintiffs do not argue that a
cost inflation methodology would have been superior; they simply argue that the agency failed to
adequately explain why it used this methodology when it had previously had concluded that a
cost inflation methodology was more accurate. However, the agency adequately explained its use
of the charge inflation methodology. Moreover, as the Court explained above, the fact that the
agency previously arrived at a different conclusion regarding the optimal inflation methodology
does not make arbitrary or capricious its subsequent conclusion that a charge inflation
methodology was preferable.
Second, Plaintiffs argue that the agency failed to account for the decline of cost-to-charge
ratios over time. However, the agency explained its reasons for, once again, declining to apply an
adjustment factor to cost-to-charge ratios. It is worth reproducing the agency’s thorough
explanation at length:
We also carefully analyzed the comments suggesting that we also adjust the cost-
to-charge ratios that are used in setting the outlier thresholds. We believe it is
necessary to inflate the charges from the FY 2004 MedPAR file to project charge
levels for FY 2006, but we do not believe it is also necessary to adjust the cost-to-
115
charge ratios from the March 2005 Provider-Specific File. The FY 2004 MedPAR
charge data include charges for dates of service through August 31, 2003.
Although these data are the most recent case-specific charge information we have
available for a complete fiscal year, the FY 2004 MedPAR charge data are over 2
years old. We likely would greatly underestimate FY 2006 outlier payments if we
did not inflate the MedPAR charge data.
On the other hand, the cost-to-charge ratios from the March 2005 Provider-
Specific File reflect much more recent hospital-specific data than the case-specific
data in the FY 2005 MedPAR file. The March 2005 Provider-Specific File
includes the cost-to-charge ratios from hospitals’ most recent tentatively-settled
cost report. In many cases, for part of FY 2006, fiscal intermediaries will
determine actual outlier payment amounts using the same cost-to-charge ratios
that are in the March 2005 Provider-Specific File. Fiscal intermediaries will begin
using an updated cost-to-charge ratio to calculate the outlier payments for a
hospital only after a more recent cost report of the hospital has been tentatively
settled. We note that the cost-to-charge ratios that we are using from the March
2005 Provider-Specific File are approximately 3 percent lower on average than
the cost-to-charge ratios from the December 2004 Provider-Specific File that we
used in setting the proposed rule outlier threshold.
70 Fed. Reg. at 47,495. This explanation adequately justifies the agency’s decision not to adjust
the cost-to-charge ratios. As the district court explained in District Hospital Partners,
“[a]lthough the Secretary’s rationale in []FY 2006 was distinct from that given in []FY 2005, it is
no less reasonable.” 973 F. Supp. 2d at 22. “Indeed, the fact that the cost-to-charge ratios used to
calculate the fixed loss threshold are actually used, for some portion of the fiscal year, to
calculate outlier payments, is a strong reason to not adjust the cost-to-charge ratios downward
based on speculation regarding the continued downward trend in cost-to-charge ratios.” Id.
Although Plaintiffs claim that the agency’s statement that lies underneath this conclusion has no
basis in the administrative record, it is actually Plaintiffs’ claim that cost-to-charge ratios would
be updated before the start of FY 2006 that is not tethered to anything in the record. Absent any
evidence to the contrary, the agency need not further justify its own understanding, as
promulgated in the Federal Register, of the complicated statutory scheme that it administers.
116
Third, Plaintiffs argue, once again, that the agency failed to account for reconciliation in
setting the fixed loss threshold. However, the agency did explain why it did not alter its
methodology to account for reconciliation:
As we did in establishing the FY 2005 outlier threshold (69 FR 49278), in our
projection of FY 2006 outlier payments we did not make an adjustment for the
possibility that hospitals’ cost-to-charge ratios and outlier payments may be
reconciled upon cost report settlement. We believe that, due to the policy
implemented in the June 9, 2003 outlier final rule, cost-to-charge ratios will no
longer fluctuate significantly and, therefore, few hospitals, if any, will actually
have these ratios reconciled upon cost report settlement. In addition, it is difficult
to predict which specific hospitals will have cost-to-charge ratios and outlier
payments reconciled in their cost reports in any given year. We also note that
reconciliation occurs because hospitals’ actual cost-to-charge ratios for the cost
reporting period are different than the interim cost-to-charge ratios used to
calculate outlier payments when a bill is processed. Our simulations assume that
cost-to-charge ratios accurately measure hospital costs and, therefore, are more
indicative of post-reconciliation than pre-reconciliation outlier payments. As a
result, we omitted any assumptions about the effects of reconciliation from the
outlier threshold calculation.
70 Fed. Reg. 47,495. This thorough explanation is certainly adequate to explain the agency’s
decision not to account for reconciliation in setting the fixed loss threshold. Indeed, Plaintiffs
point to no reason why this explanation is not adequate; they simply make the conclusory claim
that the agency’s explanation is conclusory and lacking in a factual basis. 50
In sum, the Court concludes that none of Plaintiffs’ challenges to the FY 2006 fixed loss
threshold are successful.
e. Challenge to the FY 2007 Fixed Loss Threshold Rulemaking
Lastly, the Court turns to Plaintiffs’ more modest challenges to the FY 2007 fixed loss
threshold rulemaking. Plaintiffs’ primary challenge is that, when the agency finally implemented
50
As the Court has stated above, the agency’s failure to implement reconciliation for this fiscal
year does not undermine the Court’s conclusion that it is not arbitrary and capricious for the
agency to decline to account for reconciliation explicitly.
117
an adjustment to the cost-to-charge ratios in this year, the level of the adjustment was
insufficient. Plaintiffs also argue that it was arbitrary and capricious not to account for
reconciliation. With respect to the latter challenge, for the same reasons stated beforehand with
respect to the prior fiscal years, the agency adequately explained its decision not to adjust its
methodology to account for reconciliation explicitly. See 71 Fed. Reg. at 48,149. Because the
agency’s response is similar to that offered the previous year, and because Plaintiffs have offered
no additional reasons why this response is inadequate, no further analysis is necessary for the
Court to conclude that the agency’s decision was neither arbitrary nor capricious. The Court now
turns to the adjustment factor implemented for FY 2007.
For FY 2007, the agency applied an adjustment factor of 0.9973 to the otherwise
applicable hospital-specific cost-to-charge ratios. See 71 Fed. Reg. at 48,150. To develop this
adjustment factor, the agency assessed the relationship between previous changes to costs and
previous changes to charges. See id. The agency explained thoroughly and adequately how it
developed this methodology and applied it in setting the FY 2007 fixed loss threshold. See id. In
arguing that the adjustment factor was inconsistent with the national average rates of change in
cost-to-charge ratios across previous years, Plaintiffs rely on the Exhibit 5, which they submitted
along with their motion. However, the Court previously concluded that Plaintiffs may not rely on
that exhibit because it is properly considered part of Plaintiffs’ brief and because it exceeds the
page limit previously set by the Court for briefing in this case. In any event, even if the Court
were to consider Plaintiffs’ exhibit, Plaintiffs cannot point to any “‘significant and viable and
obvious alternatives’” regarding the calculation of an adjustment factor that were before the
agency at the time of the rulemaking. Dist. Hosp. Partners, 786 F.3d at 59 (citation omitted).
Plaintiffs’ argument that the chosen adjustment factor was arbitrary and capricious is unavailing.
118
Accordingly, Plaintiffs’ claim that the FY 2007 fixed loss threshold rulemaking was arbitrary and
capricious fails.
* * *
Ultimately, while the Court concludes that it has jurisdiction over all of the claims in this
action, of all of Plaintiffs’ claims, only one is successful: the claim that the agency failed to
explain why it included the 123 identified turbo-charging hospitals in the charge inflation
calculations used to develop the FY 2004 fixed loss threshold. In all other respects, the Court
rejects Plaintiffs’ challenges to the 14 regulations at stake in this action, including Plaintiffs’
statutory challenges and Plaintiffs’ claims that the agency acted arbitrarily and capriciously in
promulgating each of those 14 regulations.
IV. CONCLUSION
For the foregoing reasons, the Court DENIES Defendant’s [126] Motion to Dismiss for
Lack of Subject Matter Jurisdiction, GRANTS IN PART and DENIES IN PART Defendant’s
[126] Motion for Summary Judgment, GRANTS IN PART and DENIES IN PART Plaintiffs’
[127/142] Motion for Summary Judgment, and GRANTS IN PART and DENIES IN PART
Plaintiffs’ [128] Motion for Judicial Notice and/or for Extra-Record Consideration of Documents
and Other Related Relief. The Court first concludes that it has subject matter jurisdiction over all
of the claims in this action. With respect to Plaintiffs’ request for the Court to consider additional
documents not part of the administrative record, the Court takes judicial notice of those publicly
available documents as necessary, but denies Plaintiffs’ request for extra-record consideration of
those documents. The Court also denies Plaintiffs’ request to add an additional comment to the
administrative record because of the untimeliness of the request. The Court denies Plaintiffs’
request to submit three additional tables and strikes from the record exhibits 5, 7, and 8 to
119
Plaintiffs’ Motion for Summary Judgment. With respect to the cross-motions for summary
judgment, the Court GRANTS Plaintiffs’ Motion and DENIES Defendant’s Motion with respect
to the claim that the agency failed to explain its decision to include 123 turbo-charging hospitals
in the dataset used to derive the inflation factor used for calculating the FY 2004 fixed loss
threshold; the Court DENIES Plaintiffs’ Motion and GRANTS Defendant’s Motion in all other
respects.
Accordingly, the FY 2004 fixed loss threshold rule is REMANDED to the agency to
provide the agency an opportunity to explain why the agency included the turbo-charging
hospitals in the data used to derive the inflation factor used to determine the FY 2004 fixed loss
threshold—or to recalculate the fixed loss threshold if necessary. In all other respects, summary
judgment is GRANTED to Defendant. The Court will retain jurisdiction pending the limited
remand to the agency regarding the FY 2004 rulemaking. See Cobell v. Norton, 240 F.3d 1081,
1109 (D.C. Cir. 2001) (district court has authority to retain jurisdiction pending remand to
agency).
An appropriate Order accompanies this Memorandum Opinion.
Dated: September 2, 2015
/s/
COLLEEN KOLLAR-KOTELLY
United States District Judge
120