When your rescue medication becomes the thing keeping the headaches coming

The relief that runs out faster every week

The sumatriptan used to hold. You'd take one, the photophobia would ease by mid-afternoon, and you'd get your evening back. Lately it clears the headache for a few hours and then the next one lands early, meaner, sitting behind the same eye. So you take another. You're up to reaching for the bottle four, five, six days a week now, and the attacks aren't spacing out. They're clustering.

That pattern has a name, and it isn't "the migraines are winning." It's a known, category-recognized phenomenon, and the frustrating part is that the medication doing it is the same one that genuinely works when you take it less often. Nobody warns you where the line is, because the line is a monthly count and no bottle prints one on the label.

The rescue-med rebound

Medication overuse headache, or MOH, is what happens when the acute drugs you use to abort attacks start driving a background of near-daily head pain of their own. It's in the International Classification of Headache Disorders (ICHD-3) as a distinct diagnosis, not a folk theory from a forum thread. The mechanism isn't fully mapped, but the clinical picture is well documented: frequent acute-medication use in someone who already has migraine can shift an episodic pattern into a chronic-daily one.

Here's the part that makes it a paradox instead of just a warning. The drug isn't defective and you aren't misusing it in any single instance. Each triptan on each bad day is a reasonable decision. It's the accumulation across the month that turns rescue into fuel. The trap is built out of individually correct choices.

Each triptan on each bad day is a reasonable decision. It's the accumulation across the month that turns rescue into fuel.

Why "treat every attack" is the advice that hurts here

The instinct most of us are trained into is simple: an attack starts, you treat it early and hard, because early triptan use aborts more attacks. That advice is correct for episodic migraine at a low frequency. It quietly becomes the wrong advice once your attack days climb, and nobody moves the goalpost for you when they do.

So the disease looks like it's escalating on its own. More attacks, more treatment, more attacks. A neurologist looking at that same curve sees a feedback loop where you see a worsening condition. The only way to tell the two apart is to know your acute-medication days per month, and most people, and most migraine apps, track the headache and not the treatment frequency at all.

Where the line actually sits

The thresholds aren't a vibe. ICHD-3 defines overuse by drug class, sustained for more than three months, in someone who already has a headache disorder. The numbers are lower than most people guess, and they're lowest for exactly the drugs migraineurs rely on most.

Triptans, ergots, opioids, and combination analgesics cross into overuse territory at ten or more days a month. Plain analgesics like acetaminophen, aspirin, and other NSAIDs sit at fifteen. Mix classes and the combined threshold drops back to ten. Ten days is not a lot when a bad month can hand you fifteen attack days on its own.

What to watch, and what to actually do

The pattern in practice looks less like a crisis and more like a slow creep you only notice in hindsight. The relief window shrinks. The morning after a treated attack, a duller headache is already waiting. The days between attacks disappear. If you counted, you'd find the count climbing month over month while your sense of "how often" lagged behind reality.

A few concrete moves, in order of when they matter:

1. Count acute-medication days, not just attack days. One day you took anything to abort or treat is one medication day, even if you only needed half a dose. 2. Track by class. Six triptan days plus five NSAID days is eleven combined, past the mixed-class line, even though neither class alone tripped its own threshold. 3. Watch the trend, not the single month. A bad month happens. Three climbing months in a row is the ICHD-3 window. 4. Bring the number to your clinician before they ask. "I've averaged twelve triptan days a month since June" is a far more useful sentence than "my migraines are worse." 5. Do not stop an overused medication abruptly on your own read of a chart. Withdrawal can spike headaches for a stretch before things improve, and some classes need a managed taper.

The count is the intervention

If the thing that separates a worsening disease from a treatable feedback loop is a number, then surfacing that number early is the whole game. Not after three months of chronic-daily pain and a hard reset, but at the point where you can still ease off with a small adjustment instead of a withdrawal.

This is the gap we kept watching people fall into: apps that log the attack, the aura, the pain score, and never once show the running tally of how many days this month you actually medicated. Postdrome keeps that rolling count by class, so the ten-day and fifteen-day lines are visible while you can still act on them, alongside the after-phase symptoms most trackers drop entirely. It won't decide your treatment. It makes the number you and your neurologist need impossible to lose track of.

One genuinely hopeful footnote: not every acute drug carries the same rebound signal. The newer gepants have not shown the classic overuse pattern the older classes do, and some are used both to abort attacks and as prevention. That's a real shift, and it's worth asking your clinician where your specific medications sit rather than assuming every rescue drug behaves the same way.