The Litigation-Back Approach to ERISA Benefit Denials: How Federal Courts Evaluate Disability, Life, and AD&D Claims — and What That Means for Yours
ERISA benefit denials turn on two axes: the strength of the insurer's process and the strength of the claimant's evidence.
Understanding how federal courts evaluate both — and building the administrative record accordingly — is the difference between winning and losing.
Three different claimants. Three different stories. Three different outcomes — decided by three different federal appeals court panels in a single month.
The first is a working professional, sidelined for years with fibromyalgia, who watches his long-term disability benefits get cut off after an annual file review. He fights all the way to the Ninth Circuit. He loses.
The second is a woman whose health insurer denies a hospital bill from emergency surgery, refuses to engage with her appeal letters, and — remarkably — ignores four separate written requests for her own claim file over more than three years. She loses in the district court. The Ninth Circuit reverses on every issue.
The third is a young store manager whose spinal injury after a skiing accident leaves him unable to lift the fifty-pound boxes his job requires. His doctor and physical therapist agree he can lift only fifteen. The insurer terminates his benefits. The Eleventh Circuit affirms.
Why did the second claimant win where the first and third lost?
The answer is something I have come to understand over years of litigating ERISA claims in federal court, and it is something I think most claimants — and many attorneys — do not fully appreciate until they have seen enough cases resolve: the outcome of an ERISA benefit denial is shaped far more by what happens during the administrative process than by what happens in the courtroom. The litigation is important. Who you have representing absolutely matters. But the case is built — or lost — before the lawsuit is ever filed.
When a federal court reviews a benefit denial under ERISA, it is evaluating two things:
How strong was the insurer's process? Did the administrator follow fair, thorough, and transparent procedures — or did it cut corners, stonewall requests, and recycle form letters?
How strong is the claimant's evidence? Did the claimant build a record of objective, well-documented evidence — or are there gaps, flawed testing, and unsupported conclusions?
The interaction between these two axes determines the outcome. Not the diagnosis. Not the sympathy of the facts. Not the severity of the claimant's condition. The record — and the process that generated it.
I call this the Litigation-Back approach to ERISA claims. You start at the end, how the case will be evaluated in federal court, and you work backward through every stage to the administrative process — every medical submission, every appeal letter, every document request, every response to a peer review — with that endpoint in mind. The administrative process is not a bureaucratic obstacle to get through before the "real" fight begins. The administrative process is the fight. What goes into the record during that process is, in most cases, all a federal court will ever see.
This article explains the framework, demonstrates it through three recent federal appeals decisions, and — I hope — gives anyone facing an ERISA benefit denial a clearer understanding of what actually drives outcomes and why experienced counsel makes such a significant difference.
The Two-Axis Framework
Most ERISA benefit plans — whether they cover disability, life insurance, accidental death and dismemberment, or long-term care — give the insurer or claims administrator "discretion" to decide claims and to interpret plan terms. When a plan does this, and most modern plans do, federal courts review the resulting denial under an abuse of discretion standard. That standard is deferential: the court asks whether the administrator's decision was reasonable, not whether the court would have reached a different conclusion. Unless, of course, your attorney knows the law so well that they are able to strip this advantage from the administrator (an example: what I did to Chevron).
But deference is not a blank check. Over the past three decades, the federal courts have developed a framework for evaluating what "reasonable" means — and that framework, once you see it clearly, reveals two independent axes of inquiry.
The doctrinal foundation
The Supreme Court's decision in Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101 (1989), established the basic structure. Two later decisions shaped how it operates in practice:
Metropolitan Life Insurance Co. v. Glenn, 554 U.S. 105 (2008), held that when an insurance company both decides claims and pays them out of its own pocket, that creates a "structural conflict of interest" courts must weigh as a factor. The conflict gets more weight when the insurer's process suggests bias and less weight when the insurer has taken steps to reduce potential bias and promote accuracy.
Abatie v. Alta Health & Life Insurance Co., 458 F.3d 955 (9th Cir. 2006) (en banc), established that abuse-of-discretion review is "tempered by skepticism" when a structural conflict exists, and that the level of skepticism depends on procedural factors visible in the record.
What flowed from these decisions is the two-axis framework: courts evaluate the strength of the administrator's process and the strength of the claimant's evidence. The interaction between the two determines the outcome. But because the standard is deferential, the interaction is not symmetrical. A strong insurer process can survive even against strong claimant evidence. A weak insurer process can fall even when the claimant's evidence is thin.
The matrix
The cleanest way to visualize the framework is as a simple matrix:
| Insurer's Process Strong |
Insurer's Process Weak |
|
|---|---|---|
| Claimant's Evidence Strong |
Contested | Claimant Wins |
| Claimant's Evidence Weak |
Claimant Loses | Contested |
One green box — a clear claimant win. One red box — a clear claimant loss. Two yellow boxes — genuinely contested.
The top-left box (both sides strong) is contested because abuse-of-discretion review is deferential. Even when a claimant has compelling evidence, the court is not re-deciding the claim from scratch. If the insurer also ran a thorough, independent process and its reviewers reached a reasonable conclusion, the deference built into the standard can sustain the denial. Strong claimant evidence is necessary but not always sufficient.
The bottom-right box (both sides weak) is contested for a different reason. When the insurer's process is weak — cookie-cutter denials, stonewalled requests, no meaningful engagement — those procedural failures can independently reverse the denial. But if the claimant's underlying evidence is also thin, winning the appeal may not win the claim. The case gets remanded, the insurer cleans up its process, and the claimant still has to prove the claim on a weak record.
This framework is not limited to abuse-of-discretion review
A common misconception is that the insurer's process matters only when the court is reviewing for abuse of discretion, and that under de novo review — where the court evaluates the merits independently — only the evidence matters.
That is a forest-for-the-trees error.
Under de novo review, the court does decide the merits independently. But it decides them on a record. And what generated that record? The insurer's process. The independent medical examinations, the peer reviews, the file reviews, the vocational analyses — these are the insurer's evidence, and they are the product of the insurer's administrative process. A sloppy process produces weak evidence. Unqualified reviewers produce vulnerable opinions. Leading questions produce challengeable conclusions. Under de novo review, the court may not penalize the insurer for procedural failures directly, but bad process produces bad evidence, and bad evidence loses under any standard.
There is another dimension here that experienced ERISA counsel understand and most claimants do not: the administrative process is the insurer's only opportunity to build its evidentiary record. Once litigation starts, the insurer generally cannot go commission new IMEs or order new peer reviews. The administrative record is what it is. This means the administrative process is not just the claimant's best chance to build a winning case — it is simultaneously the insurer's only chance to build a defense.
An attorney who understands this operates at a fundamentally different level than one who treats the administrative appeal as a formality before litigation. Understanding the insurer's procedural obligations — including the "meaningful dialogue" and full-and-fair-review requirements discussed below — means understanding when the insurer is required to show its hand, when it must respond to the claimant's evidence, and when its failure to do so creates vulnerabilities that will be visible to a federal court regardless of the standard of review. Building the claimant's record while simultaneously shaping the insurer's record — making sure the strongest evidence is in the file, making sure the insurer's procedural obligations are triggered and documented, making sure treating providers respond to peer reviews with specificity — is the core of what I mean by a litigation-back approach.
It requires understanding the medicine, the procedure, the law, and the timing — and understanding how all four interact. That understanding develops over hundreds of cases. It is the reason experienced ERISA counsel makes a material difference in outcomes, and it is the lens through which every case at my firm is evaluated from the moment it comes through the door.
The April 2026 Trilogy: Three Cases, Three Outcomes
In April 2026, three federal appeals courts decided three ERISA benefit-denial cases that, read together, map the two-axis framework with unusual clarity. Two involved long-term disability benefits. One involved a health benefit claim. The framework held across all three — and it holds equally across life insurance, accidental death, and every other ERISA-governed benefit. The evidence changes; the architecture does not.
Wallace — Insurer's Process Strong, Claimant's Evidence Weak
Jeffery Wallace was a mining engineer who began collecting long-term disability benefits in March 2017 based on a fibromyalgia diagnosis. By January 2022, after an annual file review, Hartford concluded he no longer met the plan's "Any Occupation" definition of disability and terminated benefits. Wallace sued. The Ninth Circuit affirmed Hartford's decision. Wallace v. Hartford Life and Accident Ins. Co., No. 25-2716, 2026 WL 1046667 (9th Cir. Apr. 17, 2026).
The interesting question is why. Hartford had as clear a structural conflict of interest as any case presents — it was both the insurer and the administrator. Under Glenn and Abatie, that conflict should have weighed heavily against Hartford. Yet the Ninth Circuit gave it "little weight," because Hartford's process was, in the court's words, "thorough, neutral, and independent."
Specifically, Hartford did the following:
When Wallace's updated medical records arrived without restrictions or limitations, and his treating physicians failed to respond to repeated clarification requests, Hartford arranged an in-person Independent Medical Examination by a board-certified occupational medicine physician. After Wallace appealed the termination, Hartford retained a tri-morbid panel of three additional independent board-certified physicians, each of whom reviewed the entire file independently. When Wallace's counsel submitted an additional opinion letter from his treating physician during the appeal, Hartford forwarded the letter to the reviewing physicians, who issued written addenda confirming their original conclusions. Before issuing a final determination, Hartford gave Wallace another opportunity to respond.
That is what Glenn's "active steps to reduce potential bias" looks like in operation. The court found that this procedural rigor neutralized the structural conflict and made the conclusion difficult to assail. On the evidence side, Wallace's most recent supportive note from his treating physician dated to October 2019, more than two years before termination. His updated records during the relevant review period documented no restrictions. Four independent physicians — one IME, three file reviewers — all concluded he could work.
What Wallace teaches
Wallace lands squarely in the red box: insurer's process strong, claimant's evidence weak. It illustrates why that box is so hard to escape once you are in it — and why the fight has to start much earlier, while the record is being built.
Timely medical updates. Treating providers should be sending updated chart notes, test results, and attending physician statements throughout the disability period, not just when an annual review begins. A two-year gap between the most recent supportive note and the termination decision is a gap the insurer will drive a truck through.
Responsive treating physicians. When the administrator sends questions to a treating provider, that provider's failure to respond is treated by courts as a hole in the claimant's record — not as a procedural failure of the insurer. Treating providers need to understand that their willingness to respond to insurer inquiries is part of their patient's case.
Specific functional documentation. Notes that document specific functional limitations ("patient cannot lift more than 10 pounds without exacerbation of pain"; "patient cannot sit for more than 30 minutes") carry real weight. Diagnostic notes alone ("patient continues to suffer from fibromyalgia") do not.
Aggressive use of administrative-appeal opportunities. When the insurer commissions an IME or peer review, that report enters the administrative record. The claimant's response — including treating provider rebuttals, additional records, and commentary — must enter the same record before the appeal closes. After it closes, the door is shut.
This last point cannot be overstated. ERISA review is generally limited to the administrative record. Whatever is not in the file by the time the final administrative decision issues will not be considered by a federal court. That makes the administrative appeal — not the litigation — the most consequential moment in the entire process.
Campbell — Insurer's Process Weak, and That Was Enough
If Wallace shows what the red box looks like, Campbell shows the green box. This is a case where the insurer's process was so weak that the Ninth Circuit reversed without needing to resolve the strength of the claimant's underlying evidence. Campbell v. UnitedHealthcare Ins. Co., Nos. 24-5736 & 25-758, 2026 WL 982848 (9th Cir. Apr. 13, 2026).
Campbell involved a health benefit claim — specifically, a claim for services performed by an emergency surgical assistant — rather than disability or life insurance. I include it here because it demonstrates something important: the two-axis framework is not benefit-specific. The insurer's procedural obligations, the claimant's evidentiary burden, and the court's analytical approach apply with equal force across every ERISA-governed benefit type. The evidence changes — proof of disability is different from proof of a covered medical service, which is different from proof of accidental death — but the architecture does not.
Leah Campbell submitted a claim for emergency surgical assistant services. United denied the claim, citing inadequate documentation that the services were "performed." What the Ninth Circuit found was a catalogue of procedural failures:
Cookie-cutter denial letters. United issued what the court called "a stream of cookie-cutter denial letters," repeating the same rationale in response to every appeal, without acknowledging Campbell's specific arguments.
No description of what would perfect the claim. Under the Ninth Circuit's decision in Saffon v. Wells Fargo & Co. Long Term Disability Plan, 522 F.3d 863 (9th Cir. 2008), when a plan administrator denies a claim for insufficient documentation, it must describe — in language the claimant can understand — what additional material would suffice. United never did so.
Internal documentation contradicting the denial. United had documentation in its own file confirming that the surgery had been performed with the assistance of an Emergency Surgical Assistants physician.
Failure to produce the administrative record. Most strikingly, United never produced any portion of the administrative record — including the Plan document itself — despite four written requests over more than three years.
Each of these failings violates specific procedural protections embedded in ERISA and its implementing regulations. Three are worth examining in detail because they recur across benefit types:
The "meaningful dialogue" requirement
Beginning with Booton v. Lockheed Med. Benefit Plan, 110 F.3d 1461 (9th Cir. 1997), and continuing through Saffon and Montour v. Hartford Life & Accident Ins. Co., 588 F.3d 623 (9th Cir. 2009), the Ninth Circuit has required plan administrators to engage in "meaningful dialogue" with claimants. Under Booton, "if benefits are denied in whole or in part, the reason for the denial must be stated in reasonably clear language, with specific reference to the plan provisions that form the basis for the denial; if the plan administrators believe that more information is needed to make a reasoned decision, they must ask for it." That is not optional. It is a doctrinal floor — and it applies whether the denied benefit is disability, life, AD&D, or anything else under ERISA.
The "full and fair review" requirement
Under 29 U.S.C. § 1133(2), every ERISA plan must afford a "reasonable opportunity ... for a full and fair review" of any denial. The implementing regulations at 29 C.F.R. § 2560.503-1 specify what this review must include: notice of the specific reasons for the denial, reference to the specific plan provisions on which it is based, and a description of any additional material necessary to perfect the claim with an explanation of why such material is necessary.
The right to plan documents
Under 29 U.S.C. § 1024(b)(4), administrators must furnish, on written request, the plan document, summary plan description, and similar instruments. Failure to comply can trigger penalties under 29 U.S.C. § 1132(c)(1) — up to $110 per day per violation in the court's discretion. Campbell's most consequential holding is on this point: the Ninth Circuit held that United's failure to produce the Plan document for over three years independently supported penalties, because without the Plan document Campbell "lacked a reference point against which to evaluate the validity of United's reasons for denying her claims."
That insight — that the right to plan documents is not a clerical formality but a substantive prerequisite for evaluating any denial — is going to be cited in claimant briefs for years. And it applies with equal force to a disability claimant who cannot evaluate whether the insurer correctly applied the plan's definition of disability, a life insurance beneficiary who cannot assess whether an exclusion was properly invoked, or an AD&D claimant who needs to understand how the plan defines "accidental."
What Campbell teaches
The Ninth Circuit reversed on every issue: it reversed the affirmance of the denial, vacated the denial of attorneys' fees, and vacated the refusal to impose statutory penalties.
For claimants, the practical lesson is this: when a denial letter feels like a form response that does not engage with what you actually said, that is not just frustrating. That is a documented weakness in the insurer's process — one that pushes the case from the left column of the matrix to the right — and it creates grounds for reversal independent of the strength of your underlying evidence. Save every letter. Document every request. Track the dates. The procedural record is part of the case.
For attorneys, Campbell is a reminder that the insurer's procedural obligations are not just protections for the claimant — they are opportunities to build a record of process failure that will be visible to a federal court. Every unanswered document request, every vague denial letter, every failure to explain what would perfect the claim is evidence. A litigation-back approach treats these not as annoyances but as data points being collected for a purpose.
Dunham-Zemberi — When the Insurer's Process Holds and the Claimant's Evidence Doesn't
Like Wallace, Dunham-Zemberi lands in the red box — insurer's process strong, claimant's evidence weak, denial affirmed. But where Wallace's evidence failed because his medical records were outdated and his providers were unresponsive, Dunham-Zemberi's evidence failed for a different and more instructive reason: the evidence he did submit was methodologically flawed. Dunham-Zemberi v. Lincoln Life Assurance Co. of Boston, No. 22-13316, 2026 WL 1031851 (11th Cir. Apr. 16, 2026).
Bryce Dunham-Zemberi was a Mattress Firm store manager whose job required lifting up to fifty pounds. After a 2019 skiing accident and spinal fusion surgery, he received long-term disability benefits from Lincoln. In early 2021, Lincoln terminated his benefits, concluding he no longer met the plan's definition of disabled. The Eleventh Circuit affirmed.
This case did not turn on procedural failures. Lincoln's process, by the court's account, was strong. The case turned on a substantive question: what counts as "objective medical evidence" of continued disability?
The plan defined "Proof" of disability to include "standard diagnosis, chart notes, lab findings, test results, x-rays and/or other forms of objective medical evidence." Lincoln's two physiatrist file reviewers concluded the available records did not support any restrictions or limitations. Dunham-Zemberi submitted three pieces of evidence to support continued disability:
A residual functional capacity evaluation from his physical therapist concluding his maximum lifting capacity was fifteen pounds.
A letter from his primary care physician confirming the fifteen-pound limitation.
An earlier file review Lincoln itself had commissioned in September 2020.
Each was rejected. The reasons why are the most instructive part of the case.
Why the functional capacity evaluation failed
The evaluation purported to use heart-rate monitoring as an objective measure. But the heart-rate measurement was taken before lifting commenced, not during. Without continuous monitoring during the actual lifting test, there was no objective measure of whether the reported fifteen-pound ceiling was a true physiological limit or a self-imposed stopping point. The court treated this as a fatal defect.
This is not a peripheral observation. Functional capacity evaluations are the most common substantive tool used in disability claims, and they are very often designed in ways that fail this exact test. A test that documents "the patient stopped at fifteen pounds because he reported pain" is not objective evidence of a fifteen-pound ceiling. A test that monitors heart rate, blood pressure, or oxygen saturation continuously during exertion — and that documents observable signs of physical fatigue alongside subjective reports — is.
An attorney taking a litigation-back approach would have flagged this before the evaluation was ever submitted. Knowing how federal courts evaluate FCEs — and specifically knowing that courts distinguish between self-reported limitations and objectively measured ones — changes how you work with the treating provider to design the test. That is the difference between building evidence and simply collecting it.
Why the primary care letter failed
The primary care physician's letter simply restated the conclusions from the capacity evaluation without independent clinical basis. A physician's opinion that adopts a conclusion from another report, without independent supporting evidence, inherits all of that report's weaknesses. This is a recurring pattern: a physical therapist runs a flawed evaluation, the treating physician writes a letter endorsing the result, and the file ends up with two conclusions resting on the same flawed methodology rather than two independent sources of support.
Why the earlier review failed
The earlier Lincoln-commissioned review predated December 2020. Lincoln terminated benefits in January 2021. The Eleventh Circuit held that an earlier review could not speak to functional status during a later period. Evidence has a shelf life.
What Dunham-Zemberi teaches
Three teachings worth absorbing:
First: "Objective medical evidence" requirements are taken seriously. When a plan unambiguously requires objective evidence — and most modern disability plans do, often using language nearly identical to Lincoln's — claimants cannot rely on conclusory statements, even from treating physicians. The evidence has to be genuinely objective: measurable, third-party-verifiable data.
Second: Functional capacity evaluations must be designed to generate objective data, not just narrative conclusions. The single most important question to ask of any FCE is: what does this test measure that does not depend on the patient's self-report? Continuous physiological monitoring. Observable signs of exertion. Comparison to validated norms. Test-retest reliability. If the test is essentially "the patient lifted weights until he said it hurt," it is not objective — and a federal court will say so.
Third: The burden is on the claimant. Citing Melech v. Life Insurance Co. of North America, 739 F.3d 663 (11th Cir. 2014), the Eleventh Circuit confirmed that plan language placing the burden of proof on the claimant means the administrator does not have to "ferret out evidence" the claimant did not provide. If your physical therapist says she is "available to answer questions" but provides a flawed evaluation, the burden is on you to fix the evaluation — not on the insurer to ask follow-up questions.
Reading the Trilogy Together
Returning to the matrix, now with the cases mapped:
| Insurer's Process Strong |
Insurer's Process Weak |
|
|---|---|---|
| Claimant's Evidence Strong |
Contested Deference to the administrator means strong evidence alone may not be enough |
Claimant Wins Campbell v. UnitedHealthcare Process failure reversed the denial on every issue |
| Claimant's Evidence Weak |
Claimant Loses Wallace v. Hartford Outdated records, unresponsive providers Dunham-Zemberi v. Lincoln Flawed FCE methodology, no objective support |
Contested Process failures may reverse the denial, but weak evidence complicates the remand |
Several things become clearer when the cases are examined side-by-side.
The only clear win for the claimant is green — weak insurer process, strong claimant evidence.Campbell lives here. But notice what made it green: the insurer's process was not just imperfect — it was comprehensively weak. Cookie-cutter denials, no meaningful engagement, no description of what would perfect the claim, and a three-year refusal to produce the claim file. The Ninth Circuit reversed on every issue. The teaching is not that any procedural imperfection tips the scale; it is that systematic process failure — the kind that deprives the claimant of the ability to meaningfully participate in the administrative review — independently defeats a denial regardless of the underlying evidence.
The only clear loss for the claimant is red — strong insurer process, weak claimant evidence. Both Wallace and Dunham-Zemberi land here, but for different evidentiary reasons. Wallace's evidence was weak because his medical records were outdated and his treating providers were unresponsive — a failure of record maintenance. Dunham-Zemberi's evidence was weak because the evidence he submitted was methodologically flawed — a failure of evidence design. Both failures are preventable, but they require different kinds of attention. The first is about maintaining current, specific documentation throughout the claim. The second is about ensuring that medical evaluations and provider opinions generate genuinely objective data.
Most real cases live in the yellow. The two contested boxes — where both sides are strong, or where both sides are weak — are where the majority of ERISA benefit-denial disputes actually play out. The framework's value is that it tells claimants and their attorneys where to invest effort. Building strong evidence moves the case from the bottom row to the top row. Identifying and documenting process weaknesses pushes the case from the left column to the right. The strongest position is to work both axes simultaneously: build an evidence record that can withstand scrutiny and identify and preserve every procedural failure the insurer commits along the way.
The structural conflict of interest is a factor, not a strategy. Many practitioners and claimants assume that the structural conflict — the fact that the same insurance company decides claims and pays them — is itself a powerful argument. Under Glenn and Abatie it is certainly a factor. But Wallace shows how readily that factor is neutralized when the insurer's process is strong. The conflict argument gains real traction only when paired with concrete process failures of the kind catalogued in Campbell. Standing alone, it rarely moves the needle.
The framework holds across benefit types and standards of review.Wallace and Dunham-Zemberi are disability cases. Campbell is a health benefit case. The two-axis inquiry applied identically to all three. The same framework applies to life insurance claims, accidental death and dismemberment claims, long-term care claims, and any other ERISA-governed benefit — because the procedural protections, the evidentiary standards, and the judicial review framework are structural features of ERISA itself, not features of any particular benefit type. And as discussed above, the framework holds under de novo review as well as abuse-of-discretion review, because the administrative record is generated by the same process regardless of how a court ultimately scrutinizes it.
What a Litigation-Back Approach Looks Like in Practice
For claimants who have just received an ERISA benefit denial — whether disability, life, AD&D, or any other ERISA-governed benefit — and for the providers and family members helping them, here is what I recommend:
Read the denial letter carefully. Is the rationale specific, or is it vague language about "insufficient documentation"? Does it identify the specific plan provisions on which the denial is based? Does it tell you what additional material would perfect the claim? If it does not, that is already evidence of process weakness — and it should be documented.
Make a written, dated request for the entire administrative record AND the plan document. Send it immediately. The right exists under 29 U.S.C. § 1024(b)(4). Failure to comply can trigger penalties of up to $110 per day under § 1132(c)(1), as Campbell confirms. But more importantly, you need the plan document to understand what standard the insurer applied and whether it applied it correctly. You need the claim file to see what evidence the insurer relied on and what it ignored.
Calendar the appeal deadline. ERISA plans typically allow 180 days to appeal an adverse benefit determination. Missing the deadline can foreclose the entire case.
Inventory the medical evidence in the file. What is in there? What is missing? Are there specific functional limitations documented, or only diagnoses? Is there an attending physician statement? Is the evidence current?
If the case turns on functional capacity, look hard at any FCE. Was the test designed to generate objective data, or was it based on patient self-report? Was physiological monitoring continuous during exertion, or only at rest?
Respond meaningfully on appeal. Address every reason given in the denial. Submit additional medical evidence. Get treating providers to respond in writing to specific findings in any peer review or IME. Use whatever opportunities the plan affords for further response. Every item that enters the record before the appeal closes is part of the case. Everything that does not is gone.
Document every interaction with the administrator. Save every letter, email, and voicemail. Note the date and substance of every phone call. The procedural record is part of the case — and process failures that are not documented might as well not have happened.
Consider whether to consult an ERISA attorney before the administrative appeal closes. This is the most time-sensitive recommendation on this list. Once the final administrative decision issues, the record generally closes. Counsel can be most useful before that door shuts — not after. An attorney who understands what a federal court will look for in the record can help ensure the right evidence gets in, the right arguments are made, and the insurer's procedural failures are preserved. That is what litigation-back means, and it is the reason the administrative appeal is the most consequential moment in most ERISA claims.
Frequently Asked Questions
What is the "abuse of discretion" standard in an ERISA case?
When an ERISA plan grants the administrator discretion to decide claims and interpret plan terms, federal courts review denials under the abuse-of-discretion standard rather than de novo. The court asks whether the administrator's decision was reasonable in light of the plan and the administrative record. As the cases discussed in this piece show, that inquiry has two independent dimensions: the strength of the insurer's process and the strength of the claimant's evidence. The outcome depends on where the case falls on both axes.
What is de novo review, and does this framework still apply?
Under de novo review, the court evaluates the merits independently rather than deferring to the administrator. The framework still applies because the court is evaluating evidence that was generated during the administrative process. The insurer's IMEs, peer reviews, and file reviews are the product of its process — and bad process produces weak evidence regardless of the standard of review. The administrative record is also typically the only record the court will consider, meaning the insurer cannot supplement its evidence after litigation begins.
What is a "structural conflict of interest"?
A structural conflict of interest exists when the same entity both decides claims and pays them — typically an insurance company that is both administrator and insurer. Under Glenn, the conflict is a factor courts must weigh. Its weight depends on whether the administrator has taken active procedural steps to reduce bias. As Wallace illustrates, a strong independent review process can neutralize the conflict almost entirely.
What does "objective medical evidence" mean?
Most disability plans require objective medical evidence of continued disability. As Dunham-Zemberi clarifies, that means evidence that does not depend solely on patient self-report — imaging, lab findings, range-of-motion measurements, physiological monitoring during functional testing, and similar third-party-verifiable data. Treating physician letters that summarize conclusions without independent clinical basis are typically not sufficient on their own.
What is "meaningful dialogue" under ERISA?
A line of Ninth Circuit cases beginning with Booton v. Lockheed Med. Benefit Plan, 110 F.3d 1461 (9th Cir. 1997), requires plan administrators to engage with claimant submissions, identify specific reasons for denial, and — when documentation is deemed insufficient — describe in language the claimant can understand what additional material would suffice. Cookie-cutter denial letters that ignore the claimant's arguments fail this requirement. The obligation applies across all ERISA benefit types, not just disability.
Does this framework apply to life insurance and AD&D claims?
Yes. The procedural protections under ERISA, the administrative-record rule, and the judicial review standards are structural features of the statute that apply to every ERISA-governed benefit. The evidence changes — proof of accidental death is different from proof of disability — but the two-axis framework for how courts evaluate denials remains the same. The insurer's process must be strong, and the claimant's evidence must be strong. Both axes are always in play.
Can I introduce new evidence after the administrative appeal closes?
Generally no. ERISA review is limited to the administrative record. Whatever is not in the file by the time the final administrative decision issues will not be considered by a federal court. There are narrow exceptions, but the default rule is that the record closes when the final administrative decision issues. This is why the administrative appeal — not the litigation — is the most critical stage of an ERISA claim.
How long do I have to appeal an ERISA benefit denial?
Most ERISA plans give claimants 180 days to file an administrative appeal of an adverse benefit determination. The exact deadline is in the plan and the denial letter. Missing it can foreclose the case entirely.
What are statutory penalties under § 1132(c)(1)?
Under 29 U.S.C. § 1132(c)(1), a plan administrator who fails to comply with a written request for plan documents within 30 days may be liable for up to $110 per day per violation, in the court's discretion. As Campbell held, prolonged failure to produce the plan document is independently prejudicial because the claimant cannot evaluate the validity of the denial without the plan in front of them.
Closing thoughts
ERISA's abuse-of-discretion review is deferential, and de novo review — while more searching — is still constrained by the administrative record. But the deference operates within a framework that gives claimants real ground on which to fight, if they understand where the fight is.
Wallace, Campbell, and Dunham-Zemberi together map that framework. A case's outcome depends on the interaction between the strength of the insurer's process and the strength of the claimant's evidence. Only one combination is a clear loss. Two are genuinely contested. And one — strong claimant evidence paired with a weak insurer process — is a clear win.
The practical implication is that claimants and their attorneys need to work both axes simultaneously. Build the strongest possible evidence record — current, objective, specific, responsive to the insurer's own reviewers. And at the same time, document every procedural failure the insurer commits — every stonewalled request, every form-letter denial, every failure to explain what would perfect the claim. The cases that land in the green box are the cases where the claimant did both.
That is the litigation-back approach: start from how the case will be evaluated in federal court, and work backward through every stage of the administrative process with that endpoint in mind. It is a way of thinking about ERISA claims that develops over years and hundreds of cases, and it is the foundation of every case I handle.
For anyone facing an ERISA denial — disability, life, AD&D, or otherwise — the cases confirm something I say to every client: the case is built during the administrative process, not in court. Build the record with objectivity in mind. Demand the plan document and the claim file in writing. Watch for procedural shortcuts. Use every opportunity to respond before the record closes. And if you are going to consult counsel, do it before the administrative appeal deadline — not after. Whatever fight remains for the courthouse will be shaped, and in most cases decided, by the work done before the final administrative decision issues.
This article is for informational purposes and does not constitute legal advice. Every ERISA claim turns on its specific plan language and administrative record; readers facing a benefit denial should consult counsel familiar with ERISA litigation in their circuit.
Primary cases analyzed in this article:
Wallace v. Hartford Life and Accident Ins. Co., No. 25-2716, 2026 WL 1046667 (9th Cir. Apr. 17, 2026)
Campbell v. UnitedHealthcare Ins. Co., Nos. 24-5736 & 25-758, 2026 WL 982848 (9th Cir. Apr. 13, 2026)
Dunham-Zemberi v. Lincoln Life Assurance Co. of Boston, No. 22-13316, 2026 WL 1031851 (11th Cir. Apr. 16, 2026)