Does caffeine cause migraines?

Caffeine is paradoxical. Modest doses (50–100 mg) can abort an early attack — caffeine is in several combination acute medications for this reason. High doses, sudden increases, or withdrawal can trigger attacks in some people.

How much caffeine is too much for migraine?

There's no universal number. Daily intake above ~400 mg is often associated with rebound headaches; sudden cessation after regular use can trigger withdrawal headaches within 12–24 hours. The pattern that matters most is consistency, not absolute amount.

Can caffeine treat a migraine?

Caffeine is FDA-approved in combination products (e.g., Excedrin Migraine: acetaminophen + aspirin + caffeine). It enhances absorption and adds modest analgesic effect. Frequent use of caffeine-containing acute meds counts toward medication overuse headache (MOH).

Should I quit caffeine?

Not necessarily, and abrupt cessation often makes things worse short-term. If you suspect caffeine is contributing, a gradual taper over 2–3 weeks under your clinician's guidance is the conservative approach.

Caffeine and migraine: trigger, treatment, or both

Creator: Hermly
Published: 2026-05-16T00:00:00.000Z

Caffeine has a paradoxical relationship with migraine. It can abort an early attack at modest doses — which is why caffeine appears in several acute migraine medications. It can also trigger attacks at high doses, with abrupt cessation, or with the rebound patterns of frequent use. Whether caffeine is a trigger or treatment depends entirely on dose, timing, and personal pattern.

This page lays out what’s solid in the research, why the trigger question is so individual, and how to think about caffeine without swinging between extremes.

The mechanism — vasoconstriction + adenosine

Caffeine is a non-selective adenosine receptor antagonist. Adenosine accumulates during the day and signals tiredness and mild vasodilation. Caffeine blocks that signal, producing vasoconstriction of cranial blood vessels and inhibiting some nociceptive pathways.

In migraine, this gives caffeine two opposing effects:

At modest doses (50–150 mg): vasoconstriction can interrupt the early vasodilator phase of migraine, occasionally aborting an attack.
At high doses or with withdrawal: the rebound vasodilation and altered adenosine signalling can precipitate one.

The Lipton 2017 review in Journal of Headache and Pain is the standard reference on caffeine in headache management. Their summary: caffeine is genuinely useful at low intermittent doses, risky at high regular doses.

The withdrawal headache

This is the most replicable caffeine-headache pattern. People who consume caffeine daily (especially > 200 mg/day) and abruptly stop typically experience a withdrawal headache within 12–24 hours, peaking around 36 hours, resolving over 2–9 days. The headache is usually moderate, bilateral, and accompanied by fatigue and irritability.

For someone with migraine, a withdrawal headache can trigger a full migraine attack. This is why “I always get a migraine on Saturday morning” sometimes turns out to be: weekday coffee at 8 AM, weekend coffee at 11 AM. The 3-hour difference is enough.

The fix is usually not eliminating caffeine but stabilising consumption — same amount at roughly the same time each day.

The rebound pattern in combination medications

Caffeine is in several combination acute migraine medications (Excedrin Migraine, Fioricet, Anacin). It enhances analgesic absorption and adds modest pain-relief effect. The clinical trade-off:

Effective for individual attacks.
Frequent use counts toward medication overuse headache (MOH) — taking caffeine-containing combination analgesics ≥ 10 days/month for 3+ months meets criteria.
Withdrawal from frequent use produces the same withdrawal headache pattern as dietary caffeine.

Patients sometimes find themselves in the trap of: take Excedrin for migraine → withdrawal headache the next morning when caffeine wears off → take more Excedrin → frequency rises. This is one of the canonical paths into MOH.

What the predictive research shows

The Holsteen 2020 cohort tested caffeine intake (along with other triggers) prospectively. Caffeine did not emerge as a meaningful individual-level predictor across the cohort. This is consistent with the broader finding that self-reported triggers in general don’t predict reliably (see our food trigger page).

The implication isn’t that caffeine doesn’t matter — it does for specific patterns (withdrawal, rebound, very high intake). It’s that population-level associations are weaker than the individual experience would suggest, and any caffeine-attack correlation needs to be tested at the personal level.

A practical framework (not advice)

If you suspect caffeine is contributing:

Stabilise first. Same amount, same time each day, for 2–3 weeks. If the migraine pattern doesn’t change, caffeine isn’t the lever.
If you choose to reduce. Gradual taper — drop 25 mg every 3–5 days. Abrupt cessation guarantees a withdrawal headache regardless.
Don’t treat acute attacks with caffeine-containing combination meds more than 8 days/month as a self-imposed MOH buffer.
Track it. A log of intake (rough amount + timing) plus attack days for 4 weeks tells you more than this page can.

None of the above is medical advice; it’s the practical framework supported by the published reviews.

Why Hermly doesn’t read caffeine intake

Same reason as food and other self-report triggers: the Holsteen 2020 evidence on cohort prediction is weak, and adding a daily logging field imposes meaningful friction with negligible AUC gain. If caffeine is a major personal driver, your own short-term experiment will surface it faster than an app can.

Not medical advice. Hermly is a wellness app, not FDA-cleared for diagnosis or treatment. This page is informational and should not replace conversation with your doctor. If you're experiencing a sudden, severe, or unusual headache, contact a medical professional or emergency services.