Louisiana Department of Health v. HHS ( 2021 )

Case: 20-60213 Document: 00515808162 Page: 1 Date Filed: 04/05/2021
United States Court of Appeals
for the Fifth Circuit
United States Court of Appeals
Fifth Circuit
FILED
April 5, 2021
No. 20-60213 Lyle W. Cayce
Clerk
Louisiana Department of Health,
Petitioner,
versus
United States Department of Health and Human
Services; Xavier Becerra, Secretary, U.S. Department
of Health and Human Services, in his official capacity
as Secretary of the U.S. Department of Health and
Human Services,
Respondents.
Petition for Review of the Final Determination of the United States
Department of Health & Human Services
Agency No. 15-02
Before Owen, Chief Judge, and Graves and Ho, Circuit Judges.
Per Curiam:*
The Louisiana Department of Health petitions for review of a final
decision from the Secretary of the Department of Health and Human
*
Pursuant to 5th Circuit Rule 47.5, the court has determined that this
opinion should not be published and is not precedent except under the limited
circumstances set forth in 5th Circuit Rule 47.5.4.
Case: 20-60213 Document: 00515808162 Page: 2 Date Filed: 04/05/2021
No. 20-60213
Services, via the Administrator for the Centers for Medicare and Medicaid
Services (“CMS”), denying a proposed state plan amendment for
reimbursing pharmacists’ Medicaid costs. We DENY the petition for
review.
I.
The Medicaid program, enacted as Title XIX of the Social Security
Act, is a cooperative federal-state program that provides medical assistance
to low-income individuals. See 42 U.S.C. § 1396; Atkins v. Rivera, 477 U.S.
154 (1986). The federal government and the states together finance the
program, while the states administer it. “In theory, this arrangement
incentivizes states to keep rates at efficient levels, because they share
financial responsibility for Medicaid costs with the federal government.”
Alaska Dep’t of Health & Soc. Servs. v. Ctrs. for Medicare & Medicaid Servs.,
424 F.3d 931, 935 (9th Cir. 2005). The program is voluntary but, to be eligible
for federal funds, participating states must submit a “state plan” satisfying
the Medicaid statute and rules from the Secretary of the Department of
Health and Human Services. 42 U.S.C. § 1396a.
Under the Medicaid statute, the Secretary is responsible for ensuring
that state plans meet federal requirements. See Id.; Louisiana v. U.S. Dep’t of
Health & Human Servs., 905 F.2d 877, 878 (5th Cir. 1990). The Secretary has
delegated authority to carry out federal duties under the statute to the
Administrator of CMS, an agency within the Department. § 1396a. When the
Secretary, through CMS’ Administrator, approves a state’s plan, the federal
government reimburses a percentage of the state’s Medicaid expenses. 42
U.S.C. § 1396b(a)(1). “As long as the plans meet federal requirements, the
states have considerable discretion to design and operate their individual
programs.” Louisiana, 905 F.2d at 878 (citing Lewis v. Hegstrom, 767 F.2d
1371 (9th Cir. 1985)). Accordingly, CMS, “on behalf of the Secretary, is
required to approve a state plan amendment that complies with all applicable
2
Case: 20-60213 Document: 00515808162 Page: 3 Date Filed: 04/05/2021
No. 20-60213
statutes and regulations.” La. Dep’t of Health & Hosps. v. Ctr. for Medicare &
Medicaid Servs., 346 F.3d 571, 572 (5th Cir. 2003). If the Administrator
determines that a state’s plan or amendment does not meet the federal
requirements, he or she issues a disapproval determination under 42 C.F.R.
§ 430.15(c). The state may seek administrative and judicial review of these
determinations, as Louisiana has done here. See 42 U.S.C. § 1316(a)(2), (c);
42 C.F.R. §§ 430.18, 430.60.
The regulations at issue in 2012, when Louisiana sought CMS’
approval for the state plan amendment at issue in this case, referred to two
components for reimbursements paid to pharmacies for prescription drugs: a
drug’s ingredient cost and its dispensing fee. 42 C.F.R. § 447.512(b) (2012).
Section 447.512(b) addressed how states should determine payment
methodology for certain drugs. The provision stated, in pertinent part, that:
The agency payments for brand name drugs certified in
accordance with paragraph (c) of this section and drugs other
than multiple source drugs for which a specific limit has been
established must not exceed, in the aggregate, payments levels
that the agency has determined by applying the lower of the—
(1) [Estimated Acquisition Cost (“EAC”)] plus
reasonable dispensing fees established by the
agency; or
(2) Providers’ usual and customary charges to
the general public.
42 C.F.R. § 447.512(b) (2012). So under the 2012 regulations, payments for
prescription drugs could not exceed a drug’s EAC plus the provider’s
dispensing fee. 42 C.F.R. § 447.512(b)(1) (2012). The regulations defined the
EAC as the state’s “best estimate of the price generally and currently paid
by providers for a drug marketed or sold by a particular manufacturer or
labeler in the package size of drug most frequently purchased by providers.”
Id. § 447.502 (2012). A state therefore must “determine the closest estimate
3
Case: 20-60213 Document: 00515808162 Page: 4 Date Filed: 04/05/2021
No. 20-60213
possible of the actual acquisition cost,” Louisiana, 905 F.2d at 881, 1 although
the regulations did not prohibit states from relying on an average wholesale
price (“AWP”) or an average acquisition price index in making this estimate,
see 42 C.F.R. § 502.
The regulations also establish states’ burden in persuading the
Administrator that a plan meets federal requirements. The regulations
provide that the state must “maintain and make available to [CMS], upon
request, documentary evidence to support the findings.” 42 C.F.R.
§ 447.518(c). The “documentary evidence must include data, mathematical
and statistical computations, comparisons, and any other pertinent records.”
Id. Given this burden of proof, this court has stated that a state’s compliance
with § 447.512(b)’s upper-limit categories does not necessarily amount to
compliance with the state’s burden, which is to assure CMS that its
reimbursement methodology is its best estimate of costs that pharmacists
generally and currently pay. See Louisiana, 905 F.2d at 882 (“But we do not
think, given the history of the rulemaking proceeding, that a state complies
with federal requirements merely by proving its reimbursements in a
particular category do not exceed the aggregate upper limit.”). 2
1
Shortly before Louisiana submitted its state plan amendment in 2012, CMS issued
a notice of proposed rulemaking that contemplated replacing EAC with “actual acquisition
cost,” which it defined as a state’s “determination of the actual prices paid by pharmacy
providers to acquire drug products marketed or sold by specific manufacturers.” Medicaid
Program: Covered Outpatient Drugs, 77 Fed. Reg. 5320 (proposed Feb. 2, 2012) (to be
codified at 42 C.F.R. § 447.502). CMS stated that this change would render Medicaid
reimbursements more reflective of the actual prices paid.
2
The 1987 regulations at issue in Louisiana are, in relevant part, identical to the
2012 regulations at issue in this case. Compare 42 C.F.R. § 447.301 (1987) (defining
“estimated acquisition cost” as “the [state] agency’s best estimate of the price generally
and currently paid by providers for a drug marketed or sold by a particular manufacturer or
labeler in the package size of drug most frequently purchased by providers”), with 42
C.F.R. § 447.502 (2012) (defining “estimated acquisition cost” as “the [state] agency’s
best estimate of the price generally and currently paid by providers for a drug marketed or
4
Case: 20-60213 Document: 00515808162 Page: 5 Date Filed: 04/05/2021
No. 20-60213
II.
Before 2012, Louisiana calculated the EAC of many Medicaid-
covered drugs as a percentage of the drug’s AWP. Louisiana reimbursed the
acquisition cost of most brand-name drugs at either AWP minus 13.5% or
AWP minus 15%, depending on the status of the pharmacist. The discount
reflects the fact that pharmacies typically can purchase drugs below the
wholesale price. Louisiana reimbursed pharmacies for generic drugs at the
lowest of various metrics, chiefly the provider’s “usual and customary
charge” to the public.
In 2010, Louisiana began transitioning to a different reimbursement
calculation that it said would more accurately reflect Louisiana-specific costs.
Louisiana State Plan Amendment (“SPA”) 10-13 restricted maximum
compensation for multiple source drugs to 135% of a drug’s “average
acquisition cost.” CMS approved SPA 10-13, effective February 1, 2010.
Louisiana then signaled to pharmacies that more changes were on the way.
On September 28, 2012, Louisiana submitted for CMS’ approval SPA
12-55, which defined a drug’s EAC as its “average acquisition cost,”
measured by pharmacists’ actual invoices, and without any multiplier or
percentage increase. SPA 12-55 reflected the State’s analysis of several years
of data and the advice of a private consultant. The State said that the new
reimbursement methodology was “intended to establish an accurate
pharmacy reimbursement system based on actual acquisition cost (invoice)
data and a statistically validated cost of dispensing survey.” The State
acknowledged that because SPA 12-55 set prices at the average of actual
invoices, some providers would necessarily be underpaid. But SPA 12-55
sold by a particular manufacturer or labeler in the package size of drug most frequently
purchased by providers”).
5
Case: 20-60213 Document: 00515808162 Page: 6 Date Filed: 04/05/2021
No. 20-60213
provided for a review method whereby pharmacists could ask a helpdesk for
specific variations. CMS approved SPA 12-55, effective September 5, 2012.
Consistent with this expectation, “[a]lmost immediately,” some
participating pharmacies complained to the State that the new metric would
not adequately cover their costs, and Louisiana faced political pressure to
provide a more generous reimbursement rate. The State then convened a
workgroup of “more than a dozen independent and chain pharmacists.”
On November 1, the State implemented an amended plan (SPA 12-
66), based on input from the working group, that would result in higher
payments to pharmacists. SPA 12-66 proposed an adjustment to its
prescription drug payment methodology by applying multipliers or markups
to the average acquisition cost. Specifically, the State revised its definition of
EAC as follows:
Estimated Acquisition Cost (EAC)-- the Average Acquisition
Cost (AAC) of the drug dispensed adjusted by a multiplier of
1.1 for multiple source drugs and a multiplier of 1.01 for single-
source drugs. If there is not an AAC available, the EAC is equal
to the Wholesale Acquisition Cost (WAC), as reported in the
drug pricing compendia utilized by the Department’s fiscal
intermediary. For Department defined specialty therapeutic
classes, the EAC is the Wholesale Acquisition Cost adjusted by
a multiplier of 1.05.
The State explained in the press release that it would soon provide “a markup
of 10 percent” above the average acquisition cost for generic drugs and a
markup of 1 percent for brand-name drugs, and that it would reimburse
certain classes of “specialty drugs” at their “Wholesale Average Cost (a
more generous price index) plus 5 percent.” The State also amended the
dispensing fee reimbursement for all drugs from $10.13 to $10.51.
6
Case: 20-60213 Document: 00515808162 Page: 7 Date Filed: 04/05/2021
No. 20-60213
Louisiana submitted SPA 12-66 to CMS on December 21, 2012, with
a requested effective date of November 1, 2012. 3 The State told CMS that it
had “received numerous concerns from community pharmacists, legislators
and other stakeholders” about SPA 12-55’s methodology, and that the State
had conducted a “detailed review of the cost and reimbursement data
through the information submitted by community pharmacists.”
On March 19, 2013, CMS requested additional information
supporting SPA 12-66. Specifically, CMS asked the State why it reverted
from a baseline average based on actual invoices, and how it arrived at the
specific markups. The State says that it “provided CMS with some but not
all of the analyses that it had conducted.” That data consisted of a survey of
four independent pharmacies, and an accountant’s estimate that SPA 12-66
would save $30 million compared to the AWP-based methodology in place a
few years earlier. There are over 1,000 independent and chain pharmacies
operating in Louisiana. CMS followed up with several questions further
asking the State to “explain” its arrival at the multipliers. The State
responded that the figures are “[b]ased on discussions, research, and analysis
of information submitted by providers,” but the State did not provide that
underlying data. The State also responded that it implemented SPA 12-66 in
response to legislators and participating pharmacists’ criticisms of SPA 12-
55.
In September 2014, CMS communicated to the State that it would
approve SPA 12-66’s dispensing fee reimbursements but would disapprove
SPA 12-66’s reimbursement rates for ingredient costs. In response, the State
divided SPA 12-66 into two components: SPA 12-66A referred to dispensing
3
The regulations allow states to implement their plans before CMS’ approval,
although doing so risks going without federal reimbursement if the plan is later
disapproved, as happened here. See 42 C.F.R. §§ 430.20(b); 447.256(c).
7
Case: 20-60213 Document: 00515808162 Page: 8 Date Filed: 04/05/2021
No. 20-60213
fees while SPA 12-66B referred to ingredient costs. The State then returned
to reimbursing all drugs based on average acquisition cost without any
markup.
CMS issued its decision on December 11, 2014. The CMS
Administrator concluded that the State had not shown that SPA 12-66B met
42 C.F.R. § 447.502’s EAC definition. Specifically, the Administrator
concluded that the State did not sufficiently demonstrate how it arrived at
the specific multipliers, and why it reverted to the more generous wholesale
acquisition price for specialty drugs. The Administrator thus found that SPA
12-66B did not comply with § 1902(a)(30)(A), which requires that states
have methods and procedures to assure that payment rates are consistent
with efficiency, economy, and quality of care, or with the implementing
regulations at 42 C.F.R. §§447.502 and 447.512. The Administrator
accordingly disallowed Federal Financial Participation for payments to
pharmacists based on SPA 12-66B. Those payments amounted to $26 million
over the two-year period. Had CMS approved SPA 12-66B, the federal
government would have paid 61% percent of the total, or about $16 million.
The State timely requested reconsideration of CMS’ disapproval of
SPA 12-66B. The State then submitted additional data—two declarations
and thirty-one exhibits consisting mostly of spreadsheets of pharmacist
surveys—that it had not presented to CMS in its initial petition or in its
responses to CMS’ follow-up questions.
CMS’ presiding officer first held that CMS properly disapproved SPA
12-66B. He declined to review CMS’ decision de novo, and so refused to
consider the State’s supplementary evidence as untimely. The State had
cited regulations allowing discovery in the review process, and it argued that
it had the “absolute right” to introduce new evidence on reconsideration.
The presiding officer, however, concluded that the State’s cited regulations
“must be read in the context of the overall” SPA review framework, and, in
8
Case: 20-60213 Document: 00515808162 Page: 9 Date Filed: 04/05/2021
No. 20-60213
that light, “the regulations in 430 Subpart D discuss gathering and
submitting evidence that has previously been timely submitted for CMS to
consider in its initial review of the SPA.” Accordingly, the presiding officer
considered only the evidence that the State had initially supplied regarding
the four surveyed pharmacies.
On the merits, the presiding officer mainly concluded that the State
did not satisfy 42 C.F.R. § 447.512 (2012), because the State had submitted
insufficient data explaining how it arrived at the across-the-board multipliers.
He also concluded that the State’s proposed markup impermissibly
combined the ingredient costs and dispensing fee, because the State had
acknowledged that it used the multipliers to reflect both ingredient costs and
“other costs associated with dispensing” drugs. Accordingly, the presiding
officer concluded that the State had not assured that SPA 12-66’s ingredient
reimbursement methodology represented the State’s best estimate of prices
that pharmacists generally paid in 2012.
The State timely asked the Administrator to reverse the presiding
officer’s conclusions. The Administrator agreed with the presiding officer
that the additional data should not be considered, but the Administrator also
concluded that the additional data did not merit reversal even if considered.
The Administrator decided that, on the record before CMS, the State had
not demonstrated that the proposed payment increases were consistent with
the aggregate upper payment limitations set forth in 42 C.F.R. § 447.512. He
further concluded that the State’s proposed EAC calculation did not
represent the State’s “‘best estimate of the price generally and currently paid
by providers for a drug marketed or sold by a particular manufacturer or
labeler in the package size of a drug most frequently purchased by
providers.’” See 42 C.F.R. § 447.502 (2012). The Administrator did not
endorse the presiding officer’s conclusions regarding ingredient-dispensing
cost conflation, holding instead that this rationale “was not included as a
9
Case: 20-60213 Document: 00515808162 Page: 10 Date Filed: 04/05/2021
No. 20-60213
reason in the original disapproval” and that the presiding officer’s finding
was “not pertinent to upholding the disapproval.” Last, the Administrator
concluded that, even considering the supplementary evidence, that evidence
shows that SPA 12-66B reimbursement rate would overpay more than sixty
percent of pharmacies in excess of their actual costs for multiple and single
source drugs. Accordingly, the Administrator upheld CMS’ initial decision
that SPA 12-66B does not represent the State’s best estimate of costs that
Louisiana pharmacists generally paid in 2012. Louisiana timely petitioned for
review in this court.
III.
We review the Administrator’s decision disapproving a state plan
amendment under the Administrative Procedure Act, 5 U.S.C. §§ 701–706
(2003), to ensure that the decision was not arbitrary, capricious, an abuse of
discretion, or otherwise not in accordance with law. See 5 U.S.C. § 706; La.
Dep’t of Health & Hosps., 346 F.3d at 576. We also must “defer to the
Secretary’s interpretation of Medicare legislation and its attendant
regulations—the Secretary’s interpretation of Medicare regulations is given
‘controlling weight unless it is plainly erroneous or inconsistent with the
regulation.’” Id. (quoting Harris Cty. Hosp. Dist. v. Shalala, 64 F.3d 220, 221
(5th Cir. 1995)). “If the agency’s ruling meets these standards, our belief that
an alternate interpretation is more appropriate is irrelevant.” Louisiana, 905
F.2d at 881 (citing Homan & Crimen, Inc. v. Harris, 626 F.2d 1201 (5th Cir.
1980)).
IV.
The Administrator eventually reviewed the State’s supplementary
data, so we do so as well and we need not determine whether CMS correctly
refused initially to credit this later-submitted data. The supplementary
evidence consists primarily of spreadsheets of survey data that an accountant
prepared for Louisiana in December 2012. The State also submitted the
10
Case: 20-60213 Document: 00515808162 Page: 11 Date Filed: 04/05/2021
No. 20-60213
accountant’s narrative declaration that, for certain brand-name drug groups,
34% of the surveyed pharmacies had ingredient costs that exceeded SPA 12-
66B’s reimbursement rate (average acquisition cost plus 1%). The accountant
further explained that, for generic drug groups, about 37% of the pharmacies
had ingredient costs that exceeded SPA 12-66’s reimbursement rate (average
acquisition cost plus 10%).
But as the Administrator noted, these figures do “not disrupt or
counter the original supposition when implementing SPA 12-55, which never
had an expectation that all pharmacies would have their costs reimbursed,
based on average acquisition cost and that a process was provided for that
scenario in SPA 12-55.” 4 The additional data also undercuts the State’s
argument: while that data showed that SPA 12-66B’s methodology would
underpay almost forty percent of pharmacists, the Administrator noted
conversely that SPA 12-66B overpaid “more than 60 percent of pharmacies
in excess of their actual costs for” for both generic and brand-name drugs.
While an average-based metric will necessarily result in a methodology that
underpays some pharmacists, the Administrator reasonably could conclude
that SPA 12-66B overpaid most pharmacists. Finally, none of the
supplemental data addressed specialty drugs, which SPA 12-66B reimbursed
at “Wholesale Average Cost (a more generous price index) plus 5 percent.”
The Administrator could also reasonably conclude that the State had
not carried its evidentiary burden, even with the additional data. The
regulation at 42 C.F.R. § 477.518(b)(2), consistent with § 1902(a)(30)(a) of
the Social Security Act, provides that each state must “make assurances
4
The State told CMS when it proposed SPA 12-55 that, because its methodology
represents an average cost, “prices for individual drugs may sometimes be below the cost
as experienced by individual providers,” and so “[a]djustments will be made to the [general
reimbursement rate] when the overall average has increased, which can be reported to [the
State’s accountant-consultant].”
11
Case: 20-60213 Document: 00515808162 Page: 12 Date Filed: 04/05/2021
No. 20-60213
satisfactory to CMS” that the economy, efficiency, and quality of care
requirements are met, and these assurances must be supported, on CMS’
request, by “data, mathematical and statistical computations, comparisons,
and any other pertinent records.” 42 C.F.R. § 447.518(c). While a “a state’s
EAC formula may overestimate the cost of some specific drugs,” the formula
must produce “the closest, best estimate of the price pharmacists generally
and currently pay for this category as a whole.” Louisiana, 905 F.2d at 879.
None of the data show why the State retreated from the use of actual invoices
to using an inflated multiplier. CMS reasonably could be skeptical of
Louisiana’s disclaimer of reliance on actual invoices less than two months
after the State represented that actual invoices provided the most accurate
figures. On this record, CMS’ decision is not “so implausible that it could
not be ascribed to a difference in view or the product of agency expertise.”
Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43. The Administrator’s decision thus
withstands our limited appellate review.
V.
Louisiana’s remaining arguments fare no better. It contends that, in
denying SPA 12-66B, CMS held it to the more onerous but not-yet-enacted
rule requiring states’ reimbursement methodologies represent their best
estimate of pharmacists’ actual costs. But as in Louisiana, there is nothing in
the record suggesting that the State was precluded from relying on average
acquisition costs, only that CMS concluded that SPA 12-66B’s across-the-
board multipliers did not represent the state’s best estimate of prices that
pharmacists generally paid in 2012. See Louisiana, 905 F.2d at 882 (“But
there is nothing in the Administrator’s decision here that indicates that
Louisiana would not have prevailed had it been able to prove that AWP did
provide the closest price estimate.”).
Louisiana also argues that CMS’ decision is arbitrary when compared
to its treatment of other states. The State asserts Colorado as a comparator,
12
Case: 20-60213 Document: 00515808162 Page: 13 Date Filed: 04/05/2021
No. 20-60213
because there CMS approved a similar EAC model. The State acknowledges,
however, that, contrary to Colorado’s CMS-approved rates, Louisiana
previously had said that an average acquisition cost, without any multiplier
or markup, is the State’s best estimate of costs. And Colorado’s approved
methodology applied only to rural pharmacies, and thus was more targeted
than Louisiana’s across-the-board multipliers. Further, CMS approved the
Colorado plan with the caveat that it would be phased out over a one-year
period.
Louisiana also cites CMS’ determinations from 1991 involving
Arkansas and Oklahoma which it says demonstrate that it carried its burden.
Those cases involved CMS’ disapproval of state’s plan amendments after
the states’ failure to produce any evidence supporting their proposals. See
Ark. Dep’t of Human Servs., No. 90-119, 1991 WL 634857 (DAB Aug. 22,
1991); Okla. Dep’t of Human Servs., No. 90-164, 1991 WL 634860 (DAB Aug.
13, 1991). While Louisiana’s evidence certainly surpasses Oklahoma’s and
Arkansas’s from those cases, Louisiana points to no rule that those cases set
a minimum evidentiary benchmark above which CMS is obligated to approve
a state’s plan. Such a rule would contravene states’ obligation to ensure that
their plans represent their best estimates of pharmacists’ actual costs. CMS’
conclusion that Louisiana did not meet this obligation with respect to SPA
12-66 is reasonable.
The petition for review is DENIED.
13