Back

LTD & ERISA

In an ERISA disability claim, the record you build now is the only one a court may ever see. Here is how to build it.

If you only have energy for the short version, here it is. Write down what you could not do, how long you lasted before you had to stop, how long it took you to recover, and what you missed because of it. Date every entry. Keep it up over months, not days.

Then there is the part that makes this different from every other kind of documentation I write about. Long-term disability coverage (LTD) that comes through an employer is almost always governed by a federal law called the Employee Retirement Income Security Act (ERISA). Under ERISA, the file your insurance company builds while it decides your claim usually becomes the whole case. If your claim is denied and it reaches federal court, the judge generally reads only what was already in that file when the insurer made its final decision. Evidence you gather afterward usually has nowhere to go.

So the record has to be in the file before the file closes. That means during the claim, and during the appeal. Not after.

I am not a lawyer, and nothing here is legal advice. What I know is records. This is about what to write down, and when, so that whoever opens your file is reading something more than a diagnosis code and a date.

Your claim runs on a federal law that closes the file, so what is not in it usually does not count.

If your long-term disability coverage came through your job, ERISA governs it. That one fact changes how everything works, and most people never hear the word until a denial letter arrives.

In an ordinary lawsuit, you gather your evidence after you file. In an ERISA case, generally you do not. The insurer assembles what is called the administrative record while it decides your claim, and a court reviewing that decision usually looks at that record as it stood on the day of the final denial. Nothing more.

There is a second layer, and it is worth knowing the name of. Most employer policies contain a discretionary clause, which tells a court to defer to the insurer's judgment unless that judgment was unreasonable. Under that standard the question is not whether the judge would have decided your claim the same way. The question is whether the insurer had any reasonable basis for deciding it their way. A file thick with dated, specific detail is what makes an unreasonable denial look unreasonable. A thin file lets almost anything stand. Some states have banned discretionary clauses, so the review starts fresh, but I would not assume yours is one of them without checking.

Which is why the deadlines are the entire game. When a claim is denied, the federal claims rules give you at least 180 days to appeal. The insurer then has 45 days to decide, and it can take one 45-day extension. That appeal window is your last real chance to put anything into the file, and it is longer than most people realize. Use it. A one-sentence letter that says "I appeal" starts the clock and can let the insurer issue its final denial before a single piece of your evidence arrives.

Your policy writes the definition of disabled, and it usually gets harder with time.

ERISA does not define disabled. Your policy does, and the definition is not fixed.

Most employer policies start with an own-occupation standard. The question is whether you can perform the material duties of your job, the one you actually held. Then, commonly at the 24-month mark, the definition changes on its own to an any-occupation standard. Now the question is whether you can perform any job you are reasonably qualified for by education, training, or experience. Nothing about you has to change for this to happen. The date arrives and the bar moves. Insurers often begin re-reviewing a file months before the switch, which is why a request for updated records around month eighteen is rarely routine.

That timing varies more than people expect. Two years is the common case, four years is not rare, and I have seen the change written at one year and at five. Open your policy and look for the definition of disability, then look for the words "any occupation." The date those two things meet is a date you should have written down somewhere.

There is a second clock in many policies, and for our conditions it matters more than the first one. A large number of plans limit benefits to 24 months when the disability rests primarily on self-reported symptoms, meaning symptoms that cannot be measured from the outside. Pain. Fatigue. Dizziness. Headache. Nearly every plan carries a matching 24-month cap for mental health conditions. Search your policy for "self-reported" and for "mental illness" and read whatever you find twice.

None of this is meant to frighten you. It is meant to tell you what the record is for. The documentation you keep in your first year is what has to answer a question nobody has asked you yet.

Insurers ask for objective evidence, and for our conditions that request is a trap.

Denial letters lean on one phrase more than any other. Insufficient objective evidence. It sounds neutral, and it is not.

Courts reviewing these cases have drawn a line that is worth understanding, because it explains both why the phrase gets used and where it can be answered. An insurer generally cannot demand objective proof of a condition that has no objective test, since requiring evidence that does not exist is not a standard, it is a wall. But an insurer generally can require evidence of your functional limitations, which is to say evidence of what you cannot do. Those are two different demands, and denial letters routinely blur them into one.

Here is why that blur lands so hard on us. Several of our conditions do have tests. A tilt-table study or an active stand test can confirm postural orthostatic tachycardia syndrome (POTS). A Beighton score can confirm hypermobility. Tryptase and prostaglandin panels exist for mast cell activation syndrome (MCAS), and they are frequently normal in people who clearly have it. So the record arrives with a mix of positive findings and normal ones, and both get turned against the claim. The normal results are read as proof that nothing is wrong. The positive ones are read as proof that the diagnosis is real and nothing more.

Because that is the actual gap. A positive tilt table says your heart rate does what it does when you stand up. It does not say how many minutes you can stay upright before you have to lie down, how many days a week that happens, or what it costs you afterward. No lab value measures how long you lasted.

A dated record of what you could not do measures exactly that. It is the one form of evidence about your limitations that no test in the file can supply, and it is the one form only you can create.

One good day can be used against you, so the record has to show what it cost.

Insurance companies conduct surveillance. Not on every claim, but on enough of them that you should know it happens. It can mean video shot from a parked car, a review of your public social media, an interview at your kitchen table, or an examination by a physician the insurer selects and pays. Everything gathered that way goes into the same file a judge may later read.

What surveillance captures is a slice. Forty minutes of a Tuesday. A photograph of you at your sister's wedding. A clip of you carrying one grocery bag from the car to the door. What it cannot capture is the three days after.

I want to be careful here, because there is a version of this advice that tells people to stop living, and I will not give it. Go to the wedding. Carry the bag if today is a day you can carry it. The problem was never that you had a good hour. The problem is that a good hour with nothing around it looks like the truth about you, and the only thing that puts something around it is a record you were already keeping.

So the good days go in the record too, in exactly the same format as the hard ones. A file that shows nothing but your worst days is easy to dismiss, and it is also not accurate. A file that shows the real distribution, twenty-three hard days and six manageable ones and one that was almost normal, is far harder to argue with, because it is what actually happened. And when that good day is already in your own record, dated, with the cost of it written underneath, the surveillance clip stops being a contradiction. It becomes corroboration.

That is the whole method, really. Not what you did. What it cost.

Here is what to keep track of.

All of this is built around one question. What can you do, and what does your work require? The distance between those two answers is your claim.

  • The task you could not finish, and the duty it maps to. Not "bad brain fog day." Instead, "could not hold the thread of a 40-minute meeting, lost the point twice, had to ask for it in writing afterward." Your policy will ask about the material duties of your occupation, so write in the language of duties.
  • How long you lasted before you stopped. Minutes upright. Minutes at a screen. Minutes on your feet. Duration translates more directly into employability than almost anything else you can record, because a job is not a task. A job is a task sustained.
  • How long recovery took. This is the number almost nobody writes down and the one that carries the most weight. An hour of effort that costs two days in bed is not an hour of capacity.
  • What you missed, canceled, or traded away. The shift you handed off. The appointment you moved. The dinner you left at 7:30. Missed obligations are dated, external, and difficult to reinterpret.
  • Good days, in the same format as the rest. Same fields, same detail, including the cost afterward.
  • Medications, with doses and changes. A treatment history shows someone actively working to get better. Gaps in treatment get read as evidence of a mild condition, so a documented dose change or a failed trial is doing real work in your file.
  • Severity, dated, on a scale you keep consistent. Consistency matters more than precision here. A number you assign the same way every day becomes a trend, and a trend is the one thing a normal lab result cannot rebut.
  • The date. Always the date. An undated observation is a story. A dated observation is a record.

Frequency beats intensity. Six months of ordinary entries will do more for your file than one extraordinary letter written the week the denial arrives.

None of this is you building a case out of thin air.

There is a particular kind of exhaustion that comes from feeling like you are inventing your own illness on paper. You write down that standing made your heart race and your vision gray out, and some part of you braces for the person who will say that everybody gets dizzy sometimes.

You are not inventing it. Postural orthostatic tachycardia syndrome, mast cell activation syndrome, hypermobility spectrum disorder, migraine, and myalgic encephalomyelitis have literature behind them. Researchers have measured the heart rate changes, the mediator releases, the joint laxity, the post-exertional crashes. The papers exist. Most people who live with these conditions have never been handed one.

So I keep a library of them, in plain language, free, with no account required. One paper, one neutral brief describing what the study did and what it found, including the findings that cut against the hopeful reading. I do not grade the research and I do not tell you what it means for you. That is your clinician's job and yours together.

I mention it here for one narrow reason. When you write down what happened to your body, you are describing something that has been documented in the medical literature by people who took it seriously. Knowing that will not change a single word you write. It might change how it feels to write it.

A tool, for the days when organizing is the hard part.

Everything above can be done in a notebook. I mean that. If a spiral notebook by the bed is what you will actually keep up with, use the notebook and skip this section.

I built Seen. because I could not keep up with the notebook. The tracker is a daily check-in that takes about two minutes, and it exports to a physician-ready PDF with every entry dated and every pattern charted. Your data stays on your own device. I never see it.

For an ERISA claim, the relevant part is the export. The full packet is long, and the length is the point. Six months of dated entries, medication history, and severity trends is not a summary of your claim. It is the underlying record, in a form somebody else can read. Submit it during the claim or during the appeal, while the file is still open. Attach it to the appeal letter rather than mentioning that it exists.

It is $1.99 a month, and there is a hardship path if that is not possible right now. The science library is free either way.

Let me say one last thing.

Most of us are not managing one condition. We are managing a constellation, and the constellation is the reason this is hard to prove. Each piece is individually deniable. Together they take a life apart, and no single test in your file will ever say so.

That is not a flaw in you. It is a gap between how these conditions work and what the paperwork was built to measure. The record is how you close it.

Start today, with today. Not with a reconstruction of the last two years, which will exhaust you and will read as reconstruction to anyone who opens it. Just today. What you could not do, how long you lasted, what it cost you afterward, and the date.

And then get it in the file, while the file is still open. That is the part nobody tells you.

Where these rules come from.

The claims procedure, the required disclosures, and the appeal timelines described here come from the federal regulation at 29 CFR 2560.503-1, issued under section 503 of the Employee Retirement Income Security Act, 29 U.S.C. 1133. The 180-day appeal window sits at 2560.503-1(h)(3)(i) and (h)(4). Your own policy and summary plan description govern the definition of disability, the own-occupation period, and any limitation on self-reported symptoms, and those documents control where they differ from anything general I have written. This guide describes United States law only.

More in this series: FMLA intermittent leave runs on numbers only you can count. Here's how to track them  ·  POTS has no disability listing, so your record has to do the work. Here is how to build one  ·  All the guides in one place.

Ready to start?