Written and medically reviewed by Tae-Ho Guak, MD, board-certified neurologist · 25 years, 100,000+ patients
A patient emptied her bag onto my desk last month: a bottle of 200 mg ibuprofen, a nearly finished box of acetaminophen, and a printout of a normal MRI. “None of this does anything,” she said. When someone tells me ibuprofen does nothing for their migraine, the drug itself is rarely the problem. In clinic the failure almost always traces back to one of four things — the dose was too low, it was taken too late in the attack, the stomach had already stopped absorbing it, or what is being treated is not migraine. Taken properly, ibuprofen is a real migraine drug with the same grade of evidence behind it as several prescription options. Most people have simply never taken it properly.
The dose almost nobody actually takes
The standard over-the-counter tablet is 200 mg. The dose studied in migraine trials, and the one that carries a Level A recommendation from the American Headache Society, is 400 mg. So a large number of people who believe ibuprofen failed them have been treating migraine attacks at half the tested dose — often one tablet, sometimes taken with a meal that slows it down further.
Ibuprofen is also not the only option in this class, and it is not the strongest one for migraine. Soluble or effervescent aspirin at 900–1000 mg has performed comparably to sumatriptan 50 mg for two-hour headache relief in head-to-head work. Naproxen sodium 550 mg lasts longer, which matters if your attacks run long. Diclofenac potassium powder for oral solution and celecoxib oral solution are both FDA-approved specifically for acute migraine, and both were designed around the absorption problem I will come to below. In my own practice, a patient who tells me “NSAIDs don’t work” has usually tried exactly one of these, at the lowest available dose.
One caution before anyone reaches for the bottle: more is not indefinitely better. Above roughly 400–600 mg per dose the added benefit for a migraine attack flattens out while the gastric and kidney risk keeps climbing. The fix for a failing NSAID is rarely a bigger NSAID.
The window closes faster than people think
This is the single most common self-inflicted failure I see. People wait. They want to be sure it is “a real one” before spending a tablet, so they take it ninety minutes in, once the pain is established.
By then the attack has changed character. As a migraine progresses, pain-processing neurons in the brainstem and thalamus become sensitized, and the drug is no longer treating the same physiology it would have treated at minute twenty. There is a clinical marker for this that you can feel yourself: allodynia. If your scalp is tender, if your ponytail hurts, if your glasses or earrings suddenly feel unbearable, if brushing your hair is unpleasant — sensitization has set in, and both NSAIDs and triptans work substantially less well from that point on.
The instruction I give is blunt: treat within twenty to thirty minutes of knowing it is starting, at a full dose, once. Not a half dose now and a half dose later to see how it goes. That hedging strategy reliably produces the worst of both worlds — an attack that was never properly treated and a day of medication spent anyway.
Your stomach stops working during an attack
Migraine slows gastric emptying, and it does this even in people who never feel nauseated and never vomit. A tablet that would normally be absorbed in half an hour can sit in a stalled stomach while the attack builds. Patients describe this precisely without knowing what it is: “it kicks in eventually, but only after the worst is over,” or “it works on a good day and does nothing on a bad one.”
There are practical ways around it. Liquid-filled gel capsules, effervescent or dissolved formulations, and the powder-for-solution products get into the bloodstream faster than a compressed tablet. Adding metoclopramide is the other move: it is an antiemetic, it is prokinetic so it restarts gastric emptying and improves absorption of whatever you took with it, and it has modest independent activity against migraine pain. That combination — a properly dosed NSAID plus an antiemetic, taken early — outperforms the NSAID alone often enough that I consider it the default rescue plan for anyone whose oral medication is unreliable. I have written separately about what to do when pills genuinely will not stay down, where nasal and injectable routes become the answer instead.
When the drug is not the problem — the diagnosis is
If an NSAID at a correct dose, taken early, with an antiemetic, still does nothing — and a triptan has also failed — I stop adjusting drugs and start questioning the label. A headache that resists everything is usually a headache that has not been named correctly yet, which is the thread running through nearly every difficult case I see and the reason I wrote why your headache isn’t getting better.
A few specific possibilities change the whole plan. Medication-overuse headache converts a treatable episodic pattern into a daily one, and while you are in it, acute drugs stop working almost by definition. New daily persistent headache — a headache that started on a date the person can name and never left — responds poorly to standard acute treatment. And there is one pattern worth knowing about: hemicrania continua, a continuous one-sided headache that responds completely and dramatically to indomethacin and to essentially nothing else. When a patient tells me one particular anti-inflammatory worked and no other did, that is not a coincidence to be shrugged off. It is a diagnostic finding.
The same logic applies in the other direction. If triptans have been the failure point rather than NSAIDs, the reasons are different and the next steps are different — I have gone through them in detail in why triptans stop working.
The line where NSAIDs turn against you
Everything above assumes you are using these drugs a few days a month. Past a certain frequency they stop being treatment and start being the cause. For simple analgesics — plain ibuprofen, naproxen, aspirin, acetaminophen — the threshold is fifteen or more days a month for longer than three months. For combination painkillers, triptans, ergots and opioids it is only ten. Cross those lines and the diagnosis becomes medication-overuse headache, and no adjustment to dose or timing will fix it until the overused drug comes out.
My working rule with patients is two, at most three, treatment days per week — and if you consistently need that many, the honest conclusion is that you need a preventive, not a larger tablet. The other limits are the ordinary ones: NSAIDs carry real gastrointestinal bleeding, kidney and cardiovascular risk, particularly with age, with anticoagulants, or with existing kidney disease; and the FDA advises avoiding them from twenty weeks of pregnancy onward.
The patient with the bag of pills did not need a new prescription that day. She switched to 400 mg in liquid gel form with metoclopramide, taken at the first visual flicker rather than an hour into the pain, and cut her use from most days of the week to about six days a month. Roughly a third of the people who come to me convinced that “nothing over the counter works” turn out to be in that position — not out of options, just never given the instructions that make the option work. It is worth ruling that in or out before you conclude your migraines are untreatable.
Next steps
- Free headache red-flag checklist — the warning signs that mean your headache needs evaluation now, not later.
- Talk to a specialist — if your acute treatment keeps failing, the diagnosis is worth a second look.
This article is for education and does not replace a consultation with your own physician. Medication doses must be individualized. Seek emergency care immediately for a headache that peaks within seconds to a minute, a headache with fever and a stiff neck, or a headache with weakness, confusion, vision loss or difficulty speaking.
From the textbook. Chapter 14, “Acute Treatment of Migraine,” of my clinical textbook Headache Medicine covers this topic in full for doctors, with the evidence and decision pathways behind it. About the book →
