Medication Overuse Headache: How Many Days Is Too Many?

Ten days a month for triptans, ergots, opioids and combination painkillers. Fifteen for plain paracetamol, aspirin or ibuprofen taken on their own. If you use several types and none of them individually reaches its limit, ten days a month across all of them counts as well.

Those are the lines in the international diagnostic criteria. The difference between the ten and the fifteen is the part most articles leave out, and it matters: someone taking sumatriptan twelve days a month is over the line while being told by every “15 days” article that they are fine.

The thresholds, by what is actually in the packet

The criteria are set out in ICHD-3, section 8.2, the classification headache specialists work from.

Ten or more days a month

• Triptans — sumatriptan, rizatriptan, zolmitriptan, naratriptan and the rest
• Ergotamine
• Opioids — codeine, tramadol, and anything containing them
• Combination analgesics: a painkiller plus caffeine, codeine, or an antihistamine in one tablet

Fifteen or more days a month

• Paracetamol or acetaminophen on its own
• Aspirin on its own
• Ibuprofen, naproxen or another NSAID on its own

The combination-analgesic row catches the most people out, because those products are sold over the counter and read as ordinary painkillers. If the box lists more than one active ingredient, your number is ten, not fifteen.

Mixing types does not reset the count

This is the rule almost nothing outside the criteria mentions. If you take a triptan on six days and ibuprofen on six days, neither has reached its own threshold — six is under ten, six is under fifteen — but twelve days of acute medication in a month does meet the criteria on its own.

The count that matters is days you took something for a headache, not days you took any one particular thing.

Days, not doses

One tablet or four on the same day is one day. This sounds like a technicality and is genuinely the most common counting error.

It cuts both ways. A bad attack that takes three doses to break is still a single day against your threshold, which is more forgiving than people assume. But it also means the “only when I really need it” defence does not help if the days are spread out — six single tablets on six separate days is worse, by this measure, than eighteen tablets across three days.

Being over the line is not the same as having the condition

Worth being precise about, because the phrase gets used loosely.

Medication-overuse headache requires two things at once: headache on fifteen or more days a month, and regular overuse of acute medication for more than three months, in someone who already had a headache disorder. The medication threshold on its own is one half of it.

That fifteen-day figure is the same one that appears in the definition of chronic migraine, and the overlap is not a coincidence — the two numbers behind a chronic diagnosis include the fact that around half of apparent chronic migraine reverts to episodic once an overused drug is withdrawn.

So if you are over the medication line but your headache days are well below fifteen, you do not meet the criteria. What you have is a pattern worth raising, which is a different and much less alarming statement.

The three-month requirement matters too. One rough month — a bad flare, a bereavement, a stretch of poor sleep — is not this. The criteria describe something sustained.

The causal story is less settled than it sounds

The standard framing is that frequent acute medication use causes headaches to become more frequent, in a spiral. There is real support for it: people who reduce their intake often improve.

But the direction of causation is genuinely debated. Worsening headache also causes people to reach for medication more often, which produces the same correlation with the arrow reversed. Some researchers argue the condition is overdiagnosed, and that improvement after withdrawal partly reflects regression to the mean — bad periods tend to be followed by better ones whatever you do.

None of this makes the thresholds useless. They are a good prompt for a conversation. It does mean you should be suspicious of anything that tells you your medication is definitely the cause of your headaches, including anything on this site.

“Rebound” also gets used loosely for something different — a single attack returning as one dose wears off, hours later, which is a property of the drug rather than a monthly pattern. Why your migraine comes back separates the two, because they call for different fixes.

Newer acute treatments may not carry the same risk

The gepants — rimegepant and ubrogepant, taken to stop an attack — have not been associated with medication-overuse headache in the way triptans and analgesics have. That is one of the reasons they get prescribed to people who are already using acute medication frequently.

This is a live area rather than settled fact, and it is a question for whoever prescribes for you, not something to act on from a blog post.

How to count your own days

You need one number: how many days in the last thirty you took something for a headache. Not doses, not which drug, just days.

Almost nobody can produce that number from memory. It is the single most useful thing a diary gives you, and it takes about five seconds a day to keep.

Whatever you record it in, record the day you took something and what it was — the name is what determines whether your line is ten or fifteen. Migraine Journal counts your medication days over a rolling thirty-day window and shows the total against the fifteen-day mark. If you take triptans, opioids or a combination painkiller, your own line is ten, so read that count against ten rather than the number on screen. There is also a free printable diary if you would rather do it on paper, and the rest of the tracking guides start from the same day-count.

If you are over your number

Do not stop abruptly on your own initiative, and do not take that as reassurance to carry on either.

Two specific cautions. Anything containing opioids should not be stopped suddenly without medical advice. And withdrawal, when it is done, often makes headaches worse for one to two weeks before they improve — which is exactly when an unsupervised attempt gets abandoned, leaving someone convinced it does not work.

The useful move is to take the count to your GP or neurologist. “I have taken something on nineteen of the last thirty days, mostly sumatriptan” is a sentence that changes an appointment. It opens the conversation about preventive treatment, which is the actual answer for most people in this position, rather than simply using less of what you have.

That number is one of the four a headache clinician works from, so it is worth arriving with the other three as well — what to bring to a neurologist appointment sets out the full page. And if the reason you are reaching for something so often is that it keeps not working, judging that properly across several attacks is a separate exercise worth doing before the appointment rather than in it.

---

This is background reading, not medical advice, and it cannot tell you whether it applies to you. Medication-overuse headache is diagnosed by a clinician who can see your full history. If your headaches have changed in pattern or come on suddenly and severely, that needs a doctor rather than a diary.