Opinion
20-30361
09-22-2021
Clerk Michael J. P., Plaintiff-Appellee, v. Blue Cross and Blue Shield of Texas; Energy Transfer G P L P Louisiana; Energy Transfer Partners G P L P Health and Welfare Program for Active Employees, Defendants-Appellants.
Appeal from the United States District Court for the Western District of Louisiana USDC No. 2:17-CV-00764
Before HO, OLDHAM, and WILSON, Circuit Judges.
PER CURIAM. [*]
Plaintiff Michael P.'s daughter, M.P., began experiencing mental health and behavioral issues as a teenager. For years, she received counseling and hospitalization for her anxiety, depression, and multiple suicide attempts.
This case arises out of treatment M.P. received following her fifth suicide attempt in 2016. Plaintiff disagrees with Defendants' decision to reimburse only a portion of the costs for M.P.'s months-long stay and treatment at the Menninger Clinic. The district court determined that the evidence was insufficient to support Defendants' decision to deny benefits, granted Plaintiff's motion for summary judgment, and denied Defendants' motion for summary judgment. Michael P. v. Blue Cross &Blue Shield of Tex., 459 F.Supp.3d 775, 787 (W.D. La. 2020). We reverse.
I.
In December 2015, M.P. made her fourth suicide attempt and was hospitalized for twelve days. She made a fifth attempt the next month. On January 26, 2016, M.P. was admitted to Menninger, where she participated in its "COMPASS Program."
At the time of M.P.'s admission, Plaintiff was employed by Defendant Energy Transfer Partners GP, L.P. (Energy Transfer). Energy Transfer provides benefits for its employees through a self-funded employee group health benefit plan, Defendant Energy Transfer Partners GP, L.P. Health and Welfare Program for Active Employees (the Plan). The Plan is an "employee welfare benefit plan" pursuant to the Employment Retirement Income Security Act of 1974 (ERISA). M.P. was a beneficiary of the Plan during her stay at Menninger.
While Energy Transfer is the plan administrator, it is the claim administrator, Defendant Blue Cross Blue Shield of Texas, Inc. (Blue Cross), that has "final authority to establish or construe the terms and conditions of the . . . Plan and discretion to interpret and determine benefits in accordance with the . . . Plan's provisions."
The Plan also states that "[a]ll services and supplies for which benefits are available under the Plan must be Medically Necessary as determined by the Claim Administrator." The Plan further defines "Medically Necessary" or "Medical Necessity" as those services or supplies covered under the Plan which are:
1. Essential to, consistent with, and provided for the diagnosis or the direct care and treatment of the condition, sickness, disease, injury, or bodily malfunction; and
2. Provided in accordance with and are consistent with generally accepted standards of medical practice in the United States; and
3. Not primarily for the convenience of the Participant, his Physician, Behavioral Health Practitioner, the Hospital, or the Other Provider; and
4. The most economical supplies or levels of service that are appropriate for the safe and effective treatment of the Participant. When applied to hospitalization, this further means that the Participant requires acute care as a bed patient due to the nature of the services provided or the Participant's condition, and the Participant cannot receive safe or adequate care as an outpatient.
The medical staff of the Claim Administrator shall determine whether a service or supply is Medically Necessary under the Plan and will consider the views of the state and national medical communities, the guidelines and practices of Medicare, Medicaid, or other government-financed programs, and peer reviewed literature. Although a Physician, Behavioral Health Practitioner or Professional Other Provider may have prescribed treatment, such treatment may not be Medically Necessary within this definition.
Blue Cross employed the nationally recognized Milliman Care Guidelines (MCG) to evaluate whether M.P.'s treatment was "Medically Necessary." On January 27, Blue Cross approved coverage for M.P.'s inpatient treatment at Menninger through January 31. On February 1, Menninger estimated that M.P.'s total stay would be ten days, and Blue Cross approved M.P.'s continued treatment through February 5. On February 8, Blue Cross declined coverage for treatment after February 5 on the ground that acute inpatient treatment was no longer medically necessary. M.P. nevertheless remained at Menninger until March 21.
Twice, Plaintiff and Menninger appealed the denial of benefits for these additional thirty-nine days. And twice, Blue Cross affirmed its denial.
Plaintiff then requested an independent external review. A few months later, a psychiatrist named Ragy Girgis performed the review and partially overturned Blue Cross's decision, finding that M.P.'s treatment had been medically necessary during the February 6 to February 10 period. Girgis agreed, however, that M.P.'s treatment had not been medically necessary from February 11 to March 21.
Plaintiff then filed suit to recover compensation for the treatment M.P. received after February 10. Although the district court acknowledged that "there [i]s some evidence to support [Blue Cross]'s . . . determination that M.P. no longer posed an 'imminent risk' of suicide or self-harm by the last covered date and/or that a lower level of care might have been feasible," the court granted Plaintiff summary judgment and denied Defendants' motion for the same. Michael P., 459 F.Supp.3d at 786-87.
II.
"Where a benefits plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan, . . . the reviewing court applies an abuse of discretion standard to the plan administrator's decision to deny benefits." Foster v. Principal Life Ins. Co., 920 F.3d 298, 303 (5th Cir. 2019) (quotations omitted).
"Under the abuse of discretion standard, if the plan fiduciary's decision is supported by substantial evidence and is not arbitrary and capricious, it must prevail." Id. at 304 (cleaned up). To be sure, "[p]lan administrators may not arbitrarily refuse to credit a claimant's reliable evidence." Vercher v. Alexander &Alexander Inc., 379 F.3d 222, 233 (5th Cir. 2004) (alteration in original) (quoting Black &Decker Disability Plan v. Nord, 538 U.S. 822, 834 (2003)). But "it is the plan administrator's decision that must be supported by substantial evidence." Foster, 920 F.3d at 304. So "even if an ERISA plaintiff supports his claim with substantial evidence, or even with a preponderance, he will not prevail for that reason." Id. (cleaned up).
"Substantial evidence is more than a scintilla, less than a preponderance, and is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Id. (quotations omitted). "A decision is arbitrary only if made without a rational connection between the known facts and the decision or between the found facts and the evidence." Id. (quotations omitted). "Our review of the administrator's decision need not be particularly complex or technical; it need only assure that the administrator's decision falls somewhere on a continuum of reasonableness-even if on the low end." Id. (cleaned up).
We review a district court's conclusion on whether an ERISA plan administrator abused its discretion in denying benefits de novo. Id. That is, we "review the plan administrator's decision from the same perspective as the district court." Id.
III.
Blue Cross applied the MCG to determine whether Menninger's acute inpatient care was medically necessary. Plaintiff does not contest Blue Cross's use of the MCG in general. Rather, Plaintiff contends that Blue Cross's "particular application of these guidelines in this case" does not comport with the Plan Terms. We find that Blue Cross's decisions to (A) apply the MCG's acute inpatient admission criteria and (B) deny coverage based on those criteria are supported by substantial evidence.
A.
First, Plaintiff asserts that Blue Cross should have assessed the medical necessity of M.P.'s extended Menninger stay against "residential criteria" rather than "acute inpatient criteria." This is so, Plaintiff argues, because in Texas, inpatient facilities can provide residential services, Menninger offers residential as well as inpatient services, and Menninger qualifies as a residential treatment center under the terms of the Plan. Essentially, Plaintiff claims that M.P. didn't in fact receive acute inpatient treatment at Menninger.
But there is plenty of evidence to support Blue Cross's conclusion that M.P. received acute, inpatient treatment throughout her stay. Menninger repeatedly asked for permission to provide acute, inpatient care. Menninger also used the billing code "0124" (the billing code for a private "psychiatric inpatient service that provides . . . short-term intensive treatment and stabilization to individuals experiencing acute episodes of mental illness") in its requests for reimbursement-as opposed to "1001" or "1002" (the codes for forms of psychiatric "residential treatment").
Menninger's and Plaintiff's actions during the appeals process only confirm the reasonableness of Blue Cross's determination. In turning over M.P.'s medical records, Menninger noted that the level of care provided to M.P. during her stay on "01/26/16-03/21/16" was "Psychiatric Inpatient Specialty." And in Plaintiff's request for an external IRO review, he listed "mental health inpatient services 2/6/16-3/21/16" as the "health care services that are being denied."
Second, Plaintiff argues that Blue Cross abused its discretion in applying the MCG's criteria for inpatient admission "rather than criteria dictating continued treatment or discharge." But Plaintiff does not even identify a "continued treatment or discharge" standard (MCG or otherwise) that Blue Cross should have employed instead. He just cites a couple of inapposite, out-of-circuit district court opinions. In short, Plaintiff does not explain why Blue Cross's use of the MCG's admission criteria was inappropriate, let alone why it was "arbitrary and capricious."
Third, Plaintiff asserts that Blue Cross acted in an arbitrary and capricious manner by relying solely on the MCG, rather than on "accepted standards of medical practice" more generally. Plaintiff contends that the MCG erroneously focus on the alleviation of acute symptoms rather than the treatment of the patient's underlying conditions.
For support, Plaintiff points to the Plan provision that states services are medically necessary if they are "provided for the diagnosis or the direct care and treatment of the condition, sickness, disease, injury, or bodily malfunction." But another provision requires that services be provided at "[t]he most economical . . . level[] of service." And the Plan suggests that "acute care as a bed patient" is required only when "the Participant's condition" requires it and "the Participant cannot receive safe or adequate care as an outpatient." So Plaintiff is incorrect that the Plan clearly required Blue Cross to cover treatment at an acute, inpatient level until M.P.'s underlying condition was resolved.
Plaintiff also argues that Blue Cross's use of the MCG violates the Plan's provision that "medically necessary" services be provided at a level "appropriate for the safe and effective treatment" of the participant. But as discussed below, Plaintiff has not shown why it was irrational for Blue Cross to determine that M.P. could be safely treated at a lower level of care. And that's all Blue Cross determined here-that an acute, inpatient level of care was no longer necessary to treat M.P.'s symptoms. Defendants have not suggested that M.P. no longer needed any medical treatment.
Finally, Plaintiff asserts that Blue Cross's reliance on the MCG caused Blue Cross to fail to consider whether M.P.'s "substance use disorder, combined with her impulsivity issues, affected the viability or reasonableness of treatment at lower standards of care." But as explained below, the MCG specifically take into account a patient's substance abuse, disabilities, and disorders.
B.
The MCG dictate that "Admission to Inpatient Level of Care is judged appropriate" if (1) there is a "patient risk" and (2) the treatment situation and needs are appropriate at that level (i.e., the patient's condition excludes the use of a lower level of care). Substantial evidence supports Blue Cross's conclusion that neither circumstance was present after February 10.
1.
A "patient risk" exists if the patient (a) is an imminent danger to herself (or others); (b) has a life-threatening inability to perform self-care activity; (c) has a severe disability or disorder requiring acute inpatient intervention; or (d) has a severe comorbid substance abuse disorder that must be controlled to achieve stabilization of the primary psychiatric disorder.
Plaintiff does not assert that M.P. presented an imminent danger to others.
a. First, there is substantial evidence to support Blue Cross's determination that M.P. did not present an imminent danger to herself after February 10. Both before and after February 10, M.P.'s RN reassessments indicated that M.P.'s "[d]anger to self/others" had "stabilized." Likewise, none of M.P.'s safety assessments ever identified any other safety risks. On February 6, M.P. denied having suicidal ideation or thoughts of self-harm. On February 8, M.P.'s treating physician said that M.P. "ha[d] been denying [suicidal ideation] for a while now." On February 9, M.P. requested to move to a level of less-supervised care. She was at that level of less-supervised care on both February 11 and February 15. On February 10, Menninger noted that it did not see any risk of suicide or self-harm, or identify any other safety risks. Menninger also noted that while M.P. reported that suicidal ideation was still "present," it had "decreased." The physician marked this as a "passive death wish," leaving the box for "suicidal ideation" unchecked. Later that day, M.P. reported feeling safe and denied any thoughts of self-harm or suicidal ideation. By February 11, it had been seventeen days since M.P.'s last suicide attempt-arguably enough to not be considered "[v]ery recent."
The MCG provide multiple avenues for proving that a patient is an imminent danger to herself, but Plaintiff focuses only on whether M.P. was at "[i]mminent risk for recurrence of a Suicide attempt or act of serious self Harm" due to a "[v]ery recent Suicide attempt or deliberate act of serious self Harm" and an "[a]bsence of Sufficient relief of the action's precipitants."
b. Second, there is substantial evidence to support Blue Cross's determination that M.P. did not have a life-threatening inability to perform self-care after February 10. For example, M.P. generally presented as "clean and neat" during the entirety of her stay at Menninger. On February 6, M.P. told staff that she would use Benadryl to help her fall asleep earlier and avoid having anxious thoughts throughout the night. And on February 11, M.P. reported resolving a migraine by taking Excedrin and a nap.
c. Third, there is substantial evidence to support Blue Cross's determination that M.P. lacked a severe disability or disorder requiring acute inpatient intervention after February 10. For example, throughout her stay at Menninger, M.P. was largely identified as "alert and oriented" and/or "goal directed"-with an "appropriate" thought process and "intact" memory, concentration, and attention. Records also indicate that, at least as of February 5, M.P. did not experience hallucinations, delusion, or mania.
Plaintiff claims that Blue Cross erred, but he provides no record cites for M.P.'s alleged "significant impulsivity issues." See, e.g., United States v. Rojas, 812 F.3d 382, 407 n.15 (5th Cir. 2016) (failure to include supporting record citations renders an argument inadequately briefed). And he doesn't explain how these or M.P.'s "recent prior history of drug and/or alcohol abuse" compelled a finding that M.P. had a "[s]evere disability or disorder requiring acute inpatient intervention" after February 10.
Plaintiff's record cites for M.P.'s "[e]xtreme agitation or anxiety" also fail to undermine Blue Cross's decision. For example, Plaintiff cites records showing that on March 6, M.P. stated she was having "severe" anxiety-but the same report states that M.P. said she was safe, declined to discuss why she was feeling anxious, and was given drugs for "moderate anxiety." Other records show only that M.P. had "agitation/anxiety" on February 9, and "anxiety" on February 11, 15, and 22, and March 7.
Plaintiff's best evidence is a March 8 report where M.P. revealed "high" anxiety and said that discharge planning had "increased her anxiety significantly and [that] seroquel prn is not helping." But it is odd to say that M.P.'s anxiety about remaining at Menninger meant she needed to spend more time at Menninger. Regardless, all these facts (even taken together) do not preclude a finding that M.P.'s disorder did not "requir[e] acute inpatient intervention" after February 10.
Moreover, in order for Plaintiff to show that M.P. had a "[s]evere disability or disorder requiring acute inpatient intervention," Plaintiff needs to show both that M.P. had a "[s]evere behavioral health disorder-related symptom[] or condition," and that "[p]atient management at highest nonresidential level of care ha[d] failed or [was] not feasible until acute intervention or modification [was] initiated." And substantial evidence supports Blue Cross's determination that, as of February 11, patient management at the highest nonresidential level of care had not failed and was feasible without (further) acute intervention or modification. After all, on February 8, M.P.'s own treating physician said M.P. "w[ould] probably not meet acute criteria."
Plaintiff protests that because lower levels of care had failed M.P. after her previous suicide attempts, she satisfied the requirement that "[p]atient management at highest nonresidential level of care ha[d] failed" and was entitled to acute, inpatient treatment. But by that logic, Blue Cross was required to cover acute, inpatient treatment until M.P.'s underlying conditions were completely resolved. And that is clearly contrary to the Plan's focus on using "[t]he most economical . . . level[] of service that are appropriate for the safe and effective treatment of the Participant" and reserving hospitalization for situations in which the patient "requires acute care as a bed patient due to the nature of the services provided or the [patient's] condition, and the [patient] cannot receive safe or adequate care as an outpatient."
d. Finally, there is substantial evidence to support Blue Cross's determination that, at least as of February 11, M.P. did not have a severe comorbid substance abuse disorder that had to be controlled to achieve stabilization of M.P.'s primary psychiatric disorder. For example, M.P. "did not perceive of herself as having [a] substance use problem and denied the need to be on the [chemical dependence] track." And while M.P. was assigned to the partial chemical dependence track, this was due to test results showing a "high probability of moderate to severe substance disorder" and " possible mild substance use disorder." (Emphases added). In the end, she attended just a few meetings and education classes.
To be sure, Plaintiff has produced evidence that, upon admission, M.P. presented for treatment for "substance abuse." But Menninger merely noted that M.P. had consumed two cranberry vodka drinks in the past 72 hours and that her alcohol and marijuana use "varie[d]." As for the Menninger physician's conclusory statement that an inpatient level of care had been necessary because M.P.'s profound depression had been "complicated by a serious substance abuse problem that increased the likelihood of a completed suicide," this does not establish that, as of February 11, M.P. had a "[s]evere comorbid substance use disorder" that had to be controlled in order to "achieve stabilization of [M.P.'s] primary psychiatric disorder."
2.
Under the MCG, an acute inpatient level of care was medically necessary after February 10 only if there was a patient risk and the treatment situation and needs at the acute inpatient level of care were appropriate.
The treatment situation and needs are appropriate for a given level of care if (a) the patient is unwilling to participate voluntarily and requires involuntary treatment; (b) voluntary treatment at a lower level of care is not feasible (e.g., very short-term crisis intervention or residential care is unacceptable for the patient's condition); (c) there is a need for physical restraint, seclusion, or other involuntary control; or (d) around-the-clock medical or nursing care is needed to address the patient's symptoms and initiate intervention.
Substantial evidence supports Blue Cross's determination that an acute, inpatient level of treatment was not appropriate after February 10. For example, M.P. participated voluntarily in the Menninger program. Nothing in the record suggests M.P. ever required physical restraint or seclusion after February 10. And far from requiring around-the-clock medical care, M.P. was becoming increasingly independent, as evidenced by her advancement to a less-supervised level of care on February 10-11.
Once again, Plaintiff points to the fact that less-than-acute-inpatient treatment (e.g., local hospitalizations) had failed M.P. after her first four suicide attempts. But that fact does not mean that, after fifteen days of acute inpatient treatment at Menninger, Blue Cross lacked substantial evidence to conclude that further acute inpatient treatment was not medically necessary.
Plaintiff contends that many of the characteristics Blue Cross relies on to justify ending coverage (M.P.'s alertness, low risk of self-harm, appropriate thought process, cleanliness, etc.) were present throughout her stay at Menninger-including on days where Blue Cross deemed acute, inpatient care "medically necessary." But these characteristics still constitute evidence that M.P. did not present a patient risk after February 10. If anything, the fact that these characteristics existed earlier in the process suggests that Blue Cross may have been justified in covering fewer days than it did. In any case, it was not arbitrary for Blue Cross to determine that M.P.'s risk of serious self-harm-or inability to care for herself, or need for acute inpatient care, etc.-had decreased after her first few weeks at Menninger (and as her fifth suicide attempt faded in the rearview mirror).
IV.
Plaintiff's remaining arguments are meritless.
For example, Plaintiff begins his brief by pointing to all the evidence that supports a grant of benefits. But the fact that Plaintiff's position may be supported by substantial evidence is beside the point. As this court explained in Foster, it does not matter "if an ERISA plaintiff supports his claim with substantial evidence, or even with a preponderance," because "it is the plan administrator's decision that must be supported by substantial evidence, and, if it is, the administrator's decision must prevail." 920 F.3d at 304.
Plaintiff also tries to show that Blue Cross's decision was "unreasonable" by attacking the analysis conducted by its reviewers. Even assuming "unreasonable" equates to "arbitrary and capricious," however, none of Plaintiff's criticisms is persuasive.
First, Plaintiff faults the reviewers' decision to conduct file reviews as opposed to in-person analyses. But as Plaintiff himself concedes, "[Blue Cross]'s reliance on the reviewers' opinions is not arbitrary and capricious merely because they did not personally evaluate M.P." See Anderson v. Cytec Indus., Inc., 619 F.3d 505, 515 (5th Cir. 2010) ("That [Defendant's] independent experts reviewed [Plaintiff's] records but did not examine him personally . . . does not invalidate or call into question their conclusions."). In any case, at least one reviewer consulted a treating physician, and that physician acknowledged that M.P. "w[ould] probably not meet acute criteria." In short, the fact that Plaintiff believes file reviews should be given less "weight" does not mean that Blue Cross's decision lacked substantial evidence. See Foster, 920 F.3d at 304 ("Substantial evidence is more than a scintilla, less than a preponderance, and is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion."). See also Black &Decker, 538 U.S. at 834 ("Plan administrators, of course, may not arbitrarily refuse to credit a claimant's reliable evidence, including the opinions of a treating physician. But . . . courts have no warrant to require administrators automatically to accord special weight to the opinions of a claimant's physician; nor may courts impose on plan administrators a discrete burden of explanation when they credit reliable evidence that conflicts with a treating physician's evaluation.").
Second, Plaintiff again tries to attack Defendants' decision to apply the MCG alone as opposed to "accepted standards of medical practice" more generally. But the MCG clearly constitute generally accepted standards of medical practice all by themselves. See Norfolk Cnty. Ret. Sys. v. Cmty. Health Sys., Inc., 877 F.3d 687, 690 (6th Cir. 2017) (noting that 1, 000 hospitals use the MCG "[t]o determine whether a person needs inpatient or outpatient care," and explaining that the MCG "were written and reviewed by over 100 doctors and reference 15, 000 medical sources").
Finally, Plaintiff asserts that the reviewers' conclusions were dubious in light of the medical record. But Plaintiff fails to include record cites for most of the claims in this portion of his brief. And the rest of the briefing amounts to nothing more than an argument that substantial evidence also supported Plaintiff's position. As explained, that is insufficient. See Foster, 920 F.3d at 304.
Blue Cross's decision to (eventually) approve benefits through February 10-but not afterward-was not arbitrary. For starters, Blue Cross was simply deciding whether to approve or deny specific periods of treatment-it was Menninger that decided what time periods (i.e., which dates) were the subject of the requests. As Defendants put it, "the February 11 cutoff date was not arbitrary; it was a date selected by Menninger, and . . . the denial of acute care coverage after that date is supported by substantial evidence." Moreover, Blue Cross was entitled to come to a different medical-necessity conclusion as more time passed from M.P.'s latest suicide attempt. Plaintiff contends that it "strains credulity to argue that a meager four days-from February 6, 2016 through February 10, 2016-wholly mitigated . . . concerns surrounding the recency of M.P.'s latest suicide attempt" or suicidal ideation. But of course these issues were at least somewhat more distant on February 10 than they were on February 6. Furthermore, it was not until February 8 that her treating physician said M.P. "ha[d] been denying [suicidal ideation] for a while now" and "w[ould] probably not meet acute criteria," and not until February 9 that M.P. requested to move to a less-supervised level of responsibility. These facts alone provide substantial evidence for Blue Cross's decision to approve benefits from February 6 to February 10 but not afterward. Accordingly, the fact that Plaintiff can point to some evidence that cuts against Blue Cross's decision-such as the fact that M.P. continued to have "bad days," panic attacks, and passive suicidal ideation after February 10-is insufficient. Plaintiff also argues that the fact Blue Cross ultimately approved the Menninger bills for the February 6 to February 10 time period demonstrates that Blue Cross's decision to cut off benefits on February 11 was arbitrary.
In the end, Plaintiff does not and cannot undermine what the record plainly reveals: Blue Cross did not abuse its discretion and there is substantial evidence to support its denial of benefits. The district court must therefore be reversed, notwithstanding Plaintiff's discussion of why the reviewers could (or even should) have reached a different result. Defendants' denial "may not be correct, but we cannot say that it was arbitrary." Gothard v. Metro. Life Ins. Co., 491 F.3d 246, 250 (5th Cir. 2007). We reverse with instructions to enter judgment in favor of Defendants. See id.
ANDREW S. OLDHAM, Circuit Judge, concurring:
I concur in the majority opinion, which carefully and properly applies our ERISA precedents. I write separately to highlight a couple of puzzling things about the standard of review we apply in these cases. The first is its history. The substantial-evidence standard of review we apply comes from half-century old cases about pension plans under the Labor Management Relations Act. And we've continued to apply this same standard even after the Supreme Court told us it lacked a sound justification. The second puzzling thing about our standard of review is how it compares to substantial-evidence review in administrative law cases. Even though our ERISA standard of review uses the same name, it is notably more deferential than ordinary substantial-evidence review. These two features make me wonder whether our current standard for reviewing benefit denials under ERISA is justifiable.
I.
Congress enacted the Employee Retirement Income Security Act ("ERISA") in 1974. It created the cause of action that plaintiffs like Michael P. use to challenge benefit eligibility determinations. See 29 U.S.C. § 1132(a)(1)(B). But even before ERISA, federal courts occasionally heard suits challenging denials of pension benefits. These earlier suits arose under 29 U.S.C. § 186(c)-a provision of the Labor Management Relations Act of 1947 ("LMRA") that allows employers to set up pension plans. In those cases, courts settled on a standard of judicial review that considered whether the pension's "Trustees have acted arbitrarily, capriciously or in bad faith; that is, is the decision of the Trustees supported by substantial evidence or have they made an erroneous decision on a question of law." Danti v. Lewis, 312 F.2d 345, 348 (D.C. Cir. 1962); see also Giler v. Bd. of Trustees of Sheet Metal Workers Pension Plan of S. Cal., 509 F.2d 848, 849 (9th Cir. 1974); Brune v. Morse, 475 F.2d 858, 860 n.2 (8th Cir. 1973); Miniard v. Lewis, 387 F.2d 864, 865 (D.C. Cir. 1967); Kosty v. Lewis, 319 F.2d 744, 747 (D.C. Cir. 1963).
After ERISA's enactment, we adopted this same standard wholesale to adjudicate benefit eligibility disputes under ERISA, variously referring to it as an "arbitrary and capricious" standard or "substantial evidence" standard. See, e.g., Dennard v. Richards Grp., Inc., 681 F.2d 306, 313 (5th Cir. 1982); Bayles v. Central States, Se. and Sw. Areas Pension Fund, 602 F.2d 97, 99-100 (5th Cir. 1979). We even credited the D.C. Circuit's LMRA cases for developing and establishing this standard of review. See Dennard, 681 F.2d at 314. But in Firestone Tire and Rubber Co. v. Bruch, 489 U.S. 101 (1989), the Supreme Court explained that it doesn't necessarily make sense to conflate the LMRA and ERISA standards of review:
The LMRA does not provide for judicial review of the decisions of LMRA trustees. Federal courts adopted the arbitrary and capricious standard both as a standard of review and, more importantly, as a means of asserting jurisdiction over suits under § 186(c) by beneficiaries of LMRA plans who were denied benefits by trustees. Unlike the LMRA, ERISA explicitly authorizes suits against fiduciaries and plan administrators to remedy statutory violations, including breaches of fiduciary duty and lack of compliance with benefit plans. Thus, the raison d'etre for the LMRA arbitrary and capricious standard-the need for a jurisdictional basis in suits against trustees-is not present in ERISA. Without this jurisdictional analogy, LMRA
principles offer no support for the adoption of the arbitrary and capricious standard insofar as § 1132(a)(1)(B) is concerned.Firestone, 489 U.S. 953-54 (citations omitted). Firestone held that de novo review applies to benefit eligibility disputes if the plan does not expressly vest the plan administrator with discretion over eligibility determinations. Id. at 956-57. It didn't prescribe the proper standard of review for the vast majority of plans, which (especially after Firestone) expressly give the plan administrator discretion. But it did assume in dicta that courts would review decisions under such plans for "abuse of discretion." Id. at 957.
Our court continued to apply the exact same standard of review to ERISA cases[*] after Firestone. We interpreted Firestone to hold that "[i]f the [plan] administrator or fiduciary has discretionary authority, the reviewing court should apply an abuse of discretion standard." Batchelor v. Int'l Broth. of Elec. Workers Local 861 Pension and Ret. Fund, 877 F.2d 441, 442 (5th Cir. 1989). And "the way to review a decision for abuse of discretion is to determine whether the plan committee acted arbitrarily and capriciously." Penn v. Howe-Baker Engineers, Inc., 898 F.2d 1096, 1100 n.2a (5th Cir. 1990). We thus interpreted the Court's dicta endorsing an "abuse of discretion" standard to parallel our previous formulations of "arbitrary and capricious" and "substantial evidence." So formulaic recitations like this are now common in our ERISA cases: "When reviewing for arbitrary and capricious actions resulting in an abuse of discretion, we affirm an administrator's decision if it is supported by substantial evidence." Meditrust Fin. Servs. Corp. v. Sterling Chemicals, Inc., 169 F.3d 211, 215 (5th Cir. 1999). Oddly, the upshot of it all is that we're still purporting to apply the same standard of review from 1960s LMRA cases even after the Supreme Court explained the problems with that approach in Firestone.
II.
It's not just how we got here that's strange. Equally odd is the way we apply substantial-evidence review in ERISA cases. Our ERISA cases purport to review a plan administrator's decision for "substantial evidence." But ERISA's "substantial evidence" is radically different from "substantial evidence" elsewhere in law.
Take the Supreme Court's canonical substantial-evidence case, Universal Camera Corp. v. NLRB, 340 U.S. 474 (1951). There, the Supreme Court considered the view that "if what is called 'substantial evidence' is found anywhere in the record to support conclusions of fact, the courts are . . . obliged to sustain the decision without reference to how heavily the countervailing evidence may preponderate." Id. at 481 (quotation omitted). Under such a standard, it would be "enough that the evidence supporting the [agency's] result was 'substantial' when considered by itself." Id. at 478. But the Court rejected this view, finding that substantial-evidence review requires a more holistic scope. Courts engaged in substantial-evidence review must give serious consideration to "the record as a whole," "taking into account contradictory evidence or evidence from which conflicting inferences could be drawn." Id. at 487, 490; accord Dish Network Corp. v. NLRB, 953 F.3d 370, 377-78 (5th Cir. 2020).
Our approach in ERISA cases significantly diverges from this conception of substantial-evidence review. We routinely affirm plan administrator decisions without the holistic review that Universal Camera contemplates. Under our ERISA standard, "[e]ven if an ERISA plaintiff supports her claim with substantial evidence, or even with a preponderance, he will not prevail for that reason. Rather, it is the plan administrator's decision that must be supported by substantial evidence, and, if it is, the administrator's decision must prevail." Foster v. Principal Life Ins. Co., 920 F.3d 298, 304 (5th Cir. 2019). Applying this formulation, we often decline to engage in a holistic review of the evidence, because we can readily find that there is some-"more than a scintilla" even if "less than a preponderance," ibid.-evidence that supports the administrator's decision. And once we conclude that the evidence meets this low "substantial evidence" threshold we need not consider how substantial the plaintiff's evidence is, because it doesn't matter-the administrator has carried their burden. See ibid.
Our approach stems from the understandable goal of avoiding "particularly complex or technical" inquiries into the reasonableness of plan administrator decisions. Ibid. But this means we approve plan administrator decisions as long as they "fall somewhere on a continuum of reasonableness- even if on the low end." Corry v. Liberty Life Assurance Co. of Bos., 499 F.3d 389, 398 (5th Cir. 2007) (quotation omitted). In practice, any plan administrator in any case will point to some quantum of evidence which arguably puts their decision on at least the "low end" of a reasonableness spectrum. So in almost every case, we quickly approve the administrator's decision as supported by substantial evidence, without "taking into account contradictory evidence or evidence from which conflicting inferences could be drawn." Universal Camera, 340 U.S. at 487.
It appears that we've wandered far astray. The Supreme Court warned us not to use LMRA principles to review ERISA claims. We did so anyway. And then we adopted a flavor of substantial-evidence review that bears little resemblance to one we'd use in an administrative-law case. All of this makes it particularly difficult for ERISA beneficiaries to vindicate their rights under the cause of action created by Congress. And it does so with no apparent support in law, logic, or history.
[*] 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.
[*] By "ERISA cases," I mean cases where a plan beneficiary challenges a negative benefit eligibility determination by a plan administrator, and where the plan expressly gives the plan administrator discretion over such determinations. These are the vast majority of cases we hear under 29 U.S.C. § 1132(a)(1)(B).