The migraine report your neurologist can't dismiss as guesswork.
Self-reported diaries get discounted for recall bias. A tamper-evident export shows your neurologist the data is exactly what your phone recorded, unedited.
The eight-second glance
The neurologist has your printout for maybe eight seconds. She flips it, nods, sets it down, and asks how often you'd say the attacks hit. Not what the page says. What you'd say.
That small substitution, the page swapped for your verbal estimate, is the whole problem. You logged every attack for three months. Every dose. Every fog-heavy afternoon that dragged on after the head pain eased. And at the one moment it matters, the log gets read as a prompt for the same recall conversation you were trying to replace.
The credibility gap
Every self-tracked medical log arrives with an invisible asterisk. From a PDF you printed at home, a clinician cannot tell whether the numbers are what your phone recorded in the moment or what you smoothed over last night to look tidier. So the safe read is to discount it, and discounting it means falling back on your recall, which is exactly what a diary is supposed to remove.
A tamper-evident report closes that gap. It carries a verification line that proves the file has not been altered since your phone generated it. Postdrome's tamper-evident PDF export has shipped in the iOS app. When a report is generated online, the app computes a SHA-256 hash of the PDF body and sends only that hash to the postdrome.app signing service, which returns an Ed25519 digital signature; the hash, timestamp, and signature are embedded in a dedicated verification page appended to the PDF and also stored in the PDF's metadata. A recipient can drop the PDF at postdrome.app/verify, the browser recomputes the hash from the document body and checks it against the embedded Ed25519 signature to confirm the file is authentic and unaltered, without the document content ever leaving the user's device. The point is not fancier data. It is provenance a stranger can check.
More logging is not the fix
The usual advice is to track harder. Bring pages. Log every trigger, every meal, every hour of sleep. Show up with a binder and let the volume make the case.
It rarely does. A thicker stack of self-entered numbers is still self-entered numbers, and it can read as anxious rather than authoritative. What earns a second look is not how much you recorded but whether the record can be trusted to be untouched. Volume is a quantity question. Credibility is a provenance one, and they are answered by different things.
Why prospective diaries exist at all
Prospective diaries were not invented to make patients feel busy. They exist because retrospective recall of headache frequency drifts, and both patients and clinicians know it. Ask someone to estimate a three-month attack count from memory and the number tends to compress toward the last bad week or the general dread of the condition.
That is the entire reason a dated, in-the-moment entry outranks a verbal estimate. The catch is that a home-printed diary looks, to a skeptical eye, no more fixed-in-time than the estimate it replaced. Removing the edit-after-the-fact suspicion is what lets a prospective log actually do the job it was designed for.
Recall runs short
Retrospective headache-frequency estimates drift from what prospective diaries record. Miller et al. (2020) found monthly retrospective recall questionnaires may underestimate headache frequency compared to prospective diary recording. ([source](https://pmc.ncbi.nlm.nih.gov/articles/PMC9060397/)) It is the reason a dated, in-the-moment log outranks a verbal estimate in the first place.
What to show, in order
Reading the report well is a sequence, not a data dump. Lead with the verification, because it reframes everything after it, then move from pattern to specifics.
Start with the verification line so the clinician knows they are looking at the phone's record, not a retyped summary. Then walk the attack timeline rather than the monthly total. Then the postdrome hours logged after each attack, the part almost every other export drops. Then medication timing against onset, so response can be judged. Four things, in that order, keep the appointment on the record instead of on your memory of it.
Point to this
Why it lands
The verification line, first
It tells the clinician the file is the phone's record, not a retyped or edited summary.
The attack timeline, not the monthly total
Clusters and gaps carry more clinical signal than a single count.
The postdrome hours after each attack
The 24 to 72 hour after-phase is what most logs omit and what disability conversations turn on.
Medication timing against onset
Shows whether a triptan or gepant landed early enough to judge the response.
A four-line reading order for a tamper-evident export at the appointment.
What changes when the record holds
When the report is verifiable, the conversation moves off your recall and onto the pattern. The neurologist can treat the frequency as a starting fact rather than a claim to be adjusted for. A CGRP or gepant trial gets evaluated against dated onset-to-dose timing instead of a shrugged 'it maybe helped.' A disability or accommodation letter has something dated and unaltered behind it.
None of that makes the visit adversarial. It just stops you from spending it re-litigating whether your own log counts.
Where Postdrome fits
Most trackers stop at the headache and never record the hours after it, which is precisely the stretch a fair record needs. Postdrome logs the attack and the postdrome after-phase, the fatigue, residual photophobia, neck stiffness, and brain fog that outlast the pain, then exports the whole timeline in a form a clinician can check.
The honest limit matters here, so we state it plainly. A verified export is not a claim that every entry is objectively perfect. You still logged the symptoms. What it removes is the suspicion that the numbers were tidied after the fact, and that is usually the only thing standing between your data and being taken at face value.
What the stamp does and doesn't prove
A tamper-evident export proves the file is unchanged since your phone generated it. It does not claim every entry is objectively accurate. You still logged the symptoms yourself. What it removes is the suspicion that the numbers were adjusted after the fact.
What we're watching
The open question is whether clinics accept a patient-generated verification line without their own portal in the loop. Some will read it at face value. Some will want it inside their EHR before it counts. We are tracking which specialties treat a verifiable export as sufficient on its own, and where the checkable-provenance idea meets friction it hasn't yet solved.
Keep every attack, dose, and postdrome hour on your own phone, and hand your neurologist an export they can verify. One purchase, no subscription.
Postdrome keeps your timeline on-device, logs the postdrome after-phase most apps drop at the headache, and exports it without a subscription tollgate. The record is yours, and it is verifiable.
