Agendia, INC. v. Xavier BecerraAgendia, INC. v. Xavier Becerra
FOR PUBLICATION
Before: Michelle T. Friedland and Mark J. Bennett, Circuit Judges, and Frederic Block,* District Judge.
Opinion by Judge Friedland;
Dissent by Judge Block
SUMMARY**
Medicare
The panel reversed the district court‘s summary judgment in favor of Agendia, Inc. in its action alleging that the Secretary of Health and Human Services (“HHS“) wrongfully denied its claims for reimbursement for diagnostic tests under the Medicare health insurance program.
HHS reimburses medical providers for the cost of items and services that are “reasonable and necessary” for the treatment of beneficiaries. HHS employs private contractors to process providers’ claims for reimbursement. To promote consistency in initial determinations, a contractor can issue a “local coverage determination,” which specifies whether or under what conditions that contractor will approve reimbursement for some set of items or services. Agendia‘s claims for reimbursement were denied based on a local coverage determination.
Agendia argued that the denial of reimbursement was improper because it was issued without notice and opportunity for comment. The panel held that the Medicare Act‘s notice-and-comment provision –
Agendia also argued that the Medicare Act and its implementing regulations unconstitutionally delegated regulatory authority to Medicare contractors by permitting them to issue local coverage determinations. The panel held that, because those contractors acted subordinately to the HHS officials implementing Medicare, there was no unconstitutional delegation.
District Judge Block dissented from Part III.A of the majority opinion, which addressed Agendia‘s statutory claims, and from the reversal of the district court‘s grant of summary judgment to Agendia. He would hold that the “structure” of the Medicare statute was ambiguous and did not clearly support the majority‘s conclusion. Judge Block joined in Part III.B of the majority‘s opinion, which rejected Agendia‘s constitutional, non-delegation argument.
COUNSEL
Stephanie R. Marcus (argued) and Michael S. Raab, Appellate Staff; Nicola T. Hanna, United States Attorney; Ethan P. Davis, Acting Assistant Attorney General; Civil Division, United States Department of Justice, Washington, D.C.; for Defendant-Appellant/Cross-Appellee.
Patric Hooper (argued), Hooper Lundy & Bookman PC, Los Angeles, California, for Plaintiff-Appellee/Cross-Appellant.
FRIEDLAND, Circuit Judge:
Through the Medicare health insurance program, the Department of Health and Human Services (“HHS“) reimburses medical providers for the cost of items and services that are “reasonable and necessary” for the treatment of beneficiaries. HHS employs private contractors to process providers’ claims for reimbursement, including by making initial determinations as to whether the items or services for which reimbursement is sought are reasonable and necessary. To promote consistency in initial determinations, a contractor can issue a “local coverage determination,” which specifies whether or under what conditions that contractor will approve reimbursement for some set of items or services.
Plaintiff Agendia, Inc. (“Agendia“) submitted claims for reimbursement for its diagnostic tests, which were denied based on a local coverage determination. Agendia contends that the denial was improper because the local coverage determination was issued without notice and opportunity for comment in violation of a provision of the Medicare Act—specifically,
In the alternative, Agendia suggests that the Medicare Act and its implementing regulations have unconstitutionally delegated regulatory authority to Medicare contractors by permitting them to issue local coverage determinations. We hold that, because those contractors act subordinately to the HHS officials implementing Medicare, there is no unconstitutional delegation.
I.
A.
For background, we begin with a summary of the Medicare reimbursement process. Medicare Parts A and B cover only medical items and services that are “reasonable and necessary” for the treatment of beneficiaries.
The administrative appeals process consists of up to four steps: (1) a redetermination by the MAC that originally denied the claim; (2) a review by a different contractor (known as a “qualified independent contractor“); (3) a hearing before an Administrative Law Judge (“ALJ“); and finally, (4) review by the Medicare Appeals Council (“the Council“), an adjudicatory body within HHS.
Local coverage determinations, by contrast, are issued by MACs. See
B.
Agendia is a clinical laboratory that furnishes molecular diagnostic tests to doctors treating breast cancer patients. After Agendia provided such tests for eighty-six Medicare beneficiaries in 2012 and 2013, it sought reimbursement from HHS. The MAC assigned to adjudicate claims in Agendia‘s region denied payment based on a local coverage determination the MAC had previously issued. Under that local coverage determination, certain molecular diagnostic tests—including those Agendia provided—were not reasonable and necessary.
Agendia administratively appealed. The qualified independent contractor that reviewed Agendia‘s claims agreed that payment should be denied. The reviewing ALJ, however, reversed, concluding that the diagnostic tests were reasonable and necessary, notwithstanding the local coverage determination. On its own motion, the Council overturned the ALJ‘s decision, holding that the tests were not in fact reasonable and necessary. The Council explained that there was “no reason to not apply substantial deference” to the relevant local coverage determination.
Agendia then sued the Secretary in federal district court, asserting that the denial of its reimbursement claims was improper because the process for issuing the relevant local coverage determination was unlawful for two reasons.2 First, Agendia argued
The district court rejected Agendia‘s constitutional challenge but agreed with Agendia‘s statutory argument, concluding that
II.
Although the district court remanded this case, the grant of summary judgment is a final order subject to appellate review under
III.
A.
We first turn to Agendia‘s principal argument that the process for adopting local coverage determinations requires notice and comment.
The Medicare Act requires the Secretary to follow a notice-and-comment procedure for any “rule, requirement, or other statement of policy (other than a national coverage determination) that establishes or changes a substantive legal standard governing . . . the payment for services.”
The parties agree that local coverage determinations have never undergone the
We hold that local coverage determinations are not subject to the
A local coverage determination guides the application of that legal standard in a particular claim adjudication.
Specifically, it reflects a MAC‘s view of what qualifies as reasonable and necessary, and accordingly it controls that MAC‘s claims determination. But although the agency adjudicators reviewing a MAC‘s decision must consider the local coverage determination, they are not bound by it. A qualified independent contractor, an ALJ, and the Council all ultimately must apply the statutory reasonable and necessary standard to determine whether to approve a claim.5
This understanding of the effect of local coverage determinations is consistent with our court‘s precedent. We have previously explained that the reasonable and necessary standard is independent of local coverage determinations because, if such determinations “did not exist, Medicare contractors would still have an overarching duty to deny claims for items and services that are not ‘reasonable and necessary.‘” Erringer v. Thompson, 371 F.3d 625, 631 (9th Cir. 2004) (quoting
Our conclusion is also driven by the structure of the statute. Congress created a special notice-and-comment
process for national coverage determinations, requiring HHS to post a draft on the internet and provide thirty days for public comment.
Agendia‘s arguments to the contrary are not persuasive. First, Agendia asserts that the Supreme Court‘s decision in Azar v. Allina Health Services, 139 S. Ct. 1804 (2019), compels the opposite result. In that case, the Secretary argued that a Medicare reimbursement policy adopted by HHS was exempt from the
H.R. Rep. No. 114-190, at 127 (2015). Indeed, the amendment is part of a pattern of congressional actions adding procedural requirements for local coverage determinations. See
Our dissenting colleague reads the 2016 amendment as confirming the applicability of
the Medicare Act supplied the controlling legal standard, but instead he asserted that the adoption of the policy did not require notice and comment because it was an interpretative, or “gap-filling,” rule. See id. at 1816–17. The Supreme Court rejected this argument, deciding only that the
The Court, however, explicitly left open another line of argument the Secretary could pursue in future cases: “the government might have sought to argue that the policy at issue . . . didn‘t ‘establis[h] or chang[e]’ a substantive legal standard—and so didn‘t require notice and comment under
Nor are we persuaded by Agendia‘s contention that the phrase “other than a national coverage determination” in
A local coverage determination is therefore valid without undergoing the
B.
We also reject Agendia‘s alternative theory that contractors’ ability to issue local coverage determinations reflects an unconstitutional delegation of regulatory power to private entities. See
coverage determinations that supersede any conflicting local coverage determination, see
Agendia resists this conclusion by arguing that the Secretary‘s oversight is limited. First, it highlights that “unappealed Medicare claims denials based on [local coverage determinations] and other MAC policies are final.” While true, the fact that unappealed decisions are not reviewed does not mean that the Secretary—acting through an ALJ or the Council—cannot approve, disapprove, or modify a contractor‘s determination if an appeal is brought. Cf. Adkins, 310 U.S. at 388. That a particular claimant can waive or forfeit its challenge to a contractor‘s decision does not make the contractor unaccountable to the Secretary.
Second, Agendia contends that because HHS regulations allow ALJs and the Council to invalidate a local coverage determination only in a beneficiary‘s (rather than a provider‘s) appeal, Agendia must separately appeal each reimbursement claim denied by a MAC even if each is based on the same local coverage determination. See
(requiring MACs to follow certain procedures when issuing local coverage determinations).
challenging a local coverage determination directly). Although we recognize that separate appeals are burdensome, Agendia cites no authority for the proposition that burdensome limitations on remedies in an administrative review process can create an unconstitutional delegation.
Finally, Agendia contends that consideration of local coverage determinations in litigation under the False Claims Act,
IV.
Because local coverage determinations do not require notice and comment under
REVERSED.
BLOCK, District Judge, dissenting:
Agendia has been trying to secure agency approval for its BluePrint and TargetPrint tests for almost a decade.1 In 2018, it nearly succeeded. After a hearing, an ALJ issued a detailed decision that was “fully favorable” to Agendia. But Agendia‘s victory was fleeting. The Medicare Appeals Council decided, on its own motion, to review and reverse the ALJ‘s decision. Specifically, the Council held that the favorable
Consequently, the Council‘s own statements reflect that an ALJ can be reversed for failing to follow an LCD, and thus that LCDs significantly alter the nature of appellate review in Medicare cases. See generally
Agendia‘s tests, the ALJ‘s determination that they were “reasonable and necessary“—which was supported by a detailed analysis of live physician testimony—might well have been upheld and would at least have been evaluated on its merits. Instead, the ALJ‘s factual analysis was ignored and his decision reversed due to its “inconsistency” with a purportedly nonbinding LCD.
The majority acknowledges all these facts. Nonetheless, it insists that LCDs neither “establish [nor] change a substantive legal standard” because LCDs merely “guide” and do not replace the statutory “reasonable and necessary” standard. This argument elevates form over substance. In Allina Health Servs. v. Price (Allina I), then Judge Kavanaugh explained that, “a substantive legal standard at a minimum includes a standard that creates, defines and regulates the rights, duties and powers of parties.” 863 F.3d 937, 943 (D.C. Cir. 2017) (emphasis added) (internal citations omitted). Because LCDs are binding at the initial stage of the Medicare claim adjudication process and can compel the reversal of an ALJ‘s judgment, they “define and regulate the rights” of parties even if, as the majority says, they also “guide” the application of a statutory standard. See Azar v. Allina Health Servs. (Allina II), 139 S. Ct. 1804, 1812 (2019) (“if ‘a so called policy statement is in purpose or likely effect . . .a binding rule of substantive law, . . . it ‘will be taken for what it is‘“) (quoting Guardian Fed. Sav. and Loan Ass‘n v. Fed. Sav. Loan Ins. Corp., 589 F.2d 658, 666–67 (D.C. Cir. 1978)) (emphasis added). Put another way, because LCDs bind initial claim adjudicators and “narrowly limit[]” subsequent reviewers’ discretion to weigh evidence and consider arguments, they “establish” a standard at the initial stage of review and “change” the standards applied on appellate review. Fed. Sav. Loan Ins. Corp., 589 F.2d at 666–67; accord Agendia, Inc. v. Azar,
420 F. Supp. 3d 985, 997–98 (C.D.C.A. 2019) (concluding that a standard can be “substantive [regardless of] whether it is binding or entitled to substantial deference“). See generally Change, Merriam Webster Dictionary Online, https://www.merriam-webster.com/dictionary/change (last accessed Jun. 11, 2021) (defining “change” as “to make different in some particular“). They should therefore be subject to notice and comment under
Because the majority‘s selective readings of dictionaries and abstract analysis of the Medicare statute‘s “structure” do not change the reality of the administrative proceeding below, I respectfully dissent from Part III.A of the majority opinion, which addresses Agendia‘s statutory claims, and from the reversal of the district court‘s grant of summary judgment to Agendia. I join in Part III.B of the majority‘s opinion which rejects Agendia‘s constitutional, non-delegation argument.
The majority attempts to obscure the reality of the Medicare claims process—and the practical effect LCDs had on Agendia‘s claim—in two ways.
First, the majority holds that LCDs do not “change substantive legal standards” because, notwithstanding any relevant LCDs, Medicare ALJs and the Appeals Council “ultimately must apply the statutory reasonable and necessary standard.” Citing the Oxford English Dictionary (“OED“), the majority implies that a “change” occurs only when one thing is “substituted for” or “replaced with” another. See Change, Oxford English Dictionary Online,
www.oed.com/view/Entry/30468 (last visited Jun. 11, 2021). It then reasons that, because LCDs do not “replace” the statutory standard, they do not “change” that standard within the meaning of
Second, the majority contends that its decision to exempt LCDs from
Both arguments are flawed. Neither provides more than a fig leaf for the majority‘s efforts to obscure the fact that the Council reversed an ALJ‘s decision because his opinion was “inconsistent with the LCDs in effect during the dates at issue.”
A. Definitional Arguments
The phrase “substantive legal standard” and its corollary, “change a substantive legal standard,” appear to be unique. See Allina II, 139 S. Ct. at 1814 (“the phrase ‘substantive legal standard’ . . .appears in
However, the majority‘s “ordinary meaning” analysis is neither complete nor persuasive. It considers only a single definition for the word “change,” drawn from the nonlegal Oxford English Dictionary. Cf. Cox, 963 F.3d at 920-21 (rejecting a defendant‘s proposed definition of the undefined term, “notice,” because “most standard English-language dictionary. . . definitions do not define notice in relation to audience size“) (emphasis added); see also Wisconsin Cent. Ltd. v. United States, 138 S. Ct. 2067, 2070-71 (2018) (basing “ordinary meaning” analysis on three different dictionary definitions and a prior interpretation drawn from caselaw). Significantly, the word “change” can also mean “to make different in some particular” (Merriam Webster), “to make or become different” (Cambridge Dictionary), or “to alter; . . . [and] to make different in some particular” (Black‘s Law Dictionary, 6th ed.). See Change, Merriam Webster Dictionary Online, https://www.merriam-webster.com/dictionary/change (last accessed Jun. 11, 2021); Change, Cambridge Dictionary Online, https://dictionary.cambridge.org/us/dictionary/english/change (last accessed Jun. 11, 2021); Change, Black‘s Law Dictionary (6th ed. 1990). Indeed, even the majority‘s use of the OED is suspect insofar as it refers to the definition of the term “change” listed under the heading “[s]enses relating to substitution or exchange” but ignores all the definitions under the heading “[s]enses relating to alteration,
Had the majority considered these alternative definitions, it might have concluded—as I have—that a standard can “change” even if it is not replaced root and branch. It might also have realized that grafting presumptions and deference regimes onto statutory rules substantially alters the scope of the conduct those rules cover, “making them“—and the outcomes that result from their application to real cases—“different in some particular.” Such an interpretation would be consistent with the ordinary meaning of the word “change” and a more accurate reflection of the decisive role LCDs played in the administrative proceeding below.2
Because the majority‘s definitional analysis is deficient, I reject its claim that
B. Structural Arguments
The majority‘s structural analysis ignores the plain text of the statute, its legislative history and the canons of statutory interpretation.
The lack of an explicit exemption for LCDs is, however, no obstacle for the majority, which concludes that Congress must have intended to exempt LCDs from
But even assuming that the majority is right to look beyond the text of the statute, it fails to show why its interpretation of congressional intent is the right one. Cf. Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612, 1631 (2018) (“[L]egislative history is not the law“). As Justice Gorsuch points out in Allina II,
In light of the foregoing, I agree with Justice Gorsuch that the legislative history of
governmental authority has been delegated to unrepresentative agencies“). The majority fails to show why it would “make little sense” for Congress to seek these benefits for LCDs.
Moreover, Congress‘s statements in the 1987 Report suggest that it wanted courts to determine which policies are subject to rulemaking requirements based on the “effects” those policies have on stakeholders in the Medicare system. H.R. Rep. No. 100-391(I), at 430 (1987) (“The policies affected would be all those which. . .have a significant effect on Medicare enrollees, on providers, or on the administration of the program“). Because LCDs decide coverage issues as a practical matter, they have “a significant effect” on companies like Agendia and the Medicare beneficiaries they serve, so Congress probably meant them to be subject to rulemaking requirements. At the very least, Congress did not clearly intend to exempt them from such requirements.
Finally, recent changes to the Medicare statute appear to confirm that the LCDs used to deny Agendia‘s claims should have been subject to
As to the first argument, I agree with the majority that the legislative history of the Medicare Act reflects consistent concern that LCDs and other policies are being enacted without adequate procedural safeguards. See generally H.R. Rep. No. 100-391(I), at 430 (1987). However, unlike the majority, I refuse to twist Congress‘s understandable concern into an argument against imposing further safeguards. See Amalg. Transit Union Local 1398, AFL-CIO v. Laidlaw Transit Servs., Inc. 448 F.3d 1092, 1093 (9th Cir. 2006) (en banc) (“When we interpret a statute, our purpose is always to discern the intent of Congress“) (internal quotations and citations omitted). I likewise find it peculiar that the majority relies heavily on the legislative history of the 2016 amendment to refute my “structural” analysis of the Medicare Act but refuses to engage with the history of the provision we interpret today:
The second argument is the product of a selective, outcome-driven application of interpretive canon. If the majority truly believed that Congress‘s choice to enumerate exceptions to a statute implies intent to “limit[] the statute to the [exceptions] set forth,” it would agree that Congress‘s choice to explicitly exempt NCDs—and only NCDs—from
To the extent that the majority believes—again without citation or explanation—that the interpretive principle articulated in Maretta and Johnson applies solely to the 2016 amendment, I respond that my interpretation of the amendment rests on another principle cited by the majority, namely the principle that “the implications of a statute may be altered by the implications of a later statute.” See United States v. Fausto, 484 U.S. 439, 453 (1988). Put another way, my interpretation posits that the 2016 amendment may reflect congressional intent to clarify that LCDs should no longer be considered “substantive legal standards,” thereby altering
All that said, I hesitate to draw any strong conclusions from the passage of the 2016 amendment. Unlike the majority, my analysis is not “driven by the structure of the [Medicare] statute,” but rather by that statute‘s text and legislative history. I include the “structural” analysis above not because I believe it is decisive, but simply to show that the “structure” of the Medicare Act is ambiguous and does not lead inevitably to the majority‘s conclusion. In sum, I reject the majority‘s “structural” analysis because it is not grounded in the text of the Medicare Act or its legislative history. It is also undercut by subsequent amendments to that statute. The “structure” of the Medicare statute is ambiguous and does not clearly support the majority‘s conclusion.
II.
My disagreement with the majority is fundamentally definitional. Without defining its terms or citing to the congressional record, the majority gives the phrase “substantive legal standard” a narrow construction that excludes LCDs.8 By contrast, I define the term “substantive legal standard” to include all “rules” and “statements of policy” that decide Medicare claims, impact the rights of parties in the Medicare adjudicative process, or otherwise have “a significant effect” on stakeholders in the Medicare system. See H.R. Rep. No. 100-391(I), at 430 (1987) (“The policies affected would be all those which. . .have a significant effect on Medicare enrollees, on providers, or on the administration of the program“). I believe my definition takes a more realistic view of the role LCDs played in the proceedings below than does the majority, that it shows proper respect to
Today‘s opinion is a missed opportunity. In Allina II, Justice Gorsuch opened the door to judicial interpretation of the sui generis phrase “change a substantive legal standard.” 139 S. Ct. at 1814. This Court could have taken up the Supreme Court‘s challenge and defined the term “substantive legal standard” in a realistic manner. Perhaps the Supreme Court may now decide to address this important and unresolved issue.
But for now, the majority relies on an overly narrow semantic argument and a “structural” analysis that ignores the text and history of the statute it claims to interpret. In so doing, the majority obscures the substantial effects that LCDs have on companies like Agendia and ultimately, on Medicare beneficiaries.
I respectfully dissent.
Notes
Nonetheless, we infer from this amendment that local coverage determinations were not previously subject to the