Migraine Tracker.Get the app
FOOD AND TRIGGERS

Is chocolate a migraine trigger, or an early warning sign?

Blinded trials have not shown that chocolate reliably triggers migraine. Why a craving can be an early sign of an attack, and how to check your diary.

Published by Migraine Tracker Updated 8 min read

The short answer

Blinded trials have not shown that chocolate reliably triggers migraine. Only three have tested it against a look-alike placebo, in 1974, 1991 and 1997. Two found no clear effect. The smallest, with 20 people, just missed statistical significance, and its authors read it as a sign chocolate can trigger attacks in some people. A 2020 review found the evidence insufficient and said doctors should not implicitly recommend avoiding it.

Wanting chocolate can be part of the attack. ICHD-3, the headache classification, lists cravings for certain foods among the signs of the prodrome, which can start up to 48 hours before the pain. If a craving made you eat chocolate, the attack may already have begun. Your diary can show which came first.

What the blinded trials found

Three double-blind trials have tested chocolate against a placebo, and none gives clear evidence that chocolate triggers migraine in most people. In each, neither the participants nor the people assessing them knew which sample was which.

Blinded trials of chocolate and headache
Study and participantsWhat it found
Moffett et al., 1974: 25 people with migraine with aura, 23 of them women, who said even a little chocolate brought on their attacks. Two studies, 80 sessions, with chocolate or a non-cocoa bar made to taste like itOf 80 sessions, 25 ended in a headache: 13 after chocolate only, 8 after the placebo only, and 4 in the two people who reacted to both. Only 2 of the 15 people in both studies reacted to chocolate alone each time. The authors concluded that chocolate on its own rarely brings on migraine. They ran no statistical test.
Gibb et al., 1991: 20 people with migraine who believed chocolate could bring on their attacks, given chocolate or a matched placeboA typical migraine followed in 5 of 12 people given chocolate and 0 of 8 given the placebo, a median of 22 hours later. The difference just missed statistical significance (p = 0.051). The authors read it as evidence that chocolate can provoke attacks in some people who believe they are sensitive.
Marcus et al., 1997: 63 women with chronic headache, about half of them with migraine, given chocolate or carob after 2 weeks on a restricted dietChocolate was no more likely than carob to bring on a headache, whatever the type of headache and whether or not the women believed chocolate was a trigger.

The three blinded chocolate trials, from 1974, 1991 and 1997, are small and decades old, and two of them recruited people who already blamed chocolate. A PubMed search on 17 October 2026 found no newer blinded trials. A 2020 review in the journal Nutrients looked at these three trials and 25 studies of how often people report chocolate as a trigger. It concluded that there is “insufficient evidence that chocolate is a migraine trigger; thus, doctors should not make implicit recommendations to migraine patients to avoid it.”

Why chocolate gets the blame

Chocolate can get the blame even if it does not cause attacks: many people eat it most days, and a craving before an attack can lead people to eat it. Studies of named triggers cannot show what causes attacks, and in most of them only a small minority named chocolate. A 2026 meta-analysis in the Journal of Nutrition and Metabolism pooled 7 studies of self-reported triggers and found that 13.9% of people with migraine reported chocolate as a trigger (95% CI 7.0 to 21.0%). The studies differed widely, one was in children, and the review’s single author concluded that routine avoidance “cannot be universally recommended”.

Habit may explain part of the link. In a 2021 study in Frontiers in Neurology, 46% of 106 people with migraine and 47% of 88 people without migraine ate chocolate every day. Among the people with migraine, eating chocolate the day before an attack went together with reporting a food craving. If you eat chocolate most days, it will often come before an attack by chance.

A craving can be an early sign

A craving for chocolate can be an early sign of an attack rather than its cause, because food cravings are part of the prodrome. ICHD-3 describes the prodrome, also called the premonitory phase, as a phase of up to 48 hours before the pain, or before the aura in migraine with aura. Its common symptoms include tiredness, changes in mood, unusual hunger and cravings for certain foods. The NHS lists “craving certain foods or feeling thirsty” among the symptoms you may get at the start of a migraine attack, and says early symptoms can start “hours or up to 2 days before the head pain starts” (Information from the NHS website, as at 201026: Migraine). The phase after the pain is covered in our post on the migraine hangover, or postdrome.

  • How common cravings are: a 2022 meta-analysis in The Journal of Headache and Pain found that in clinic-based studies, about 11% of people with migraine reported food cravings before attacks (95% CI 3 to 32%), against 49% for tiredness, the most common early symptom. The authors said the evidence was too limited for reliable estimates.
  • Early signs predicted attacks in a diary study: in a 2003 electronic diary study in Neurology, 97 people who had non-headache symptoms in at least two of three attacks recorded them as they happened. 72% of diary entries with early warning symptoms were followed by a migraine headache. Feeling tired was the most common early symptom.
  • Named triggers and cravings go together: in a 2021 Journal of Neurology study of 53 people, those who named foods as a trigger tended to have food cravings before their attacks. The authors suggested that some reported triggers may be early signs of an attack that has already started in the brain.
  • In one person, the brain changed before the pain: in a 2016 study in the journal Brain, one person had a brain scan every morning for 30 days, covering three attacks. In the 24 hours before the pain began, the hypothalamus responded differently to a pain stimulus. It is one person, so it shows what can happen, not how often.

The Migraine Trust, a UK charity, gives chocolate as its example: a craving in the prodrome is satisfied, and chocolate then gets the blame. It does not recommend cutting out foods unless you are sure a specific food triggers your migraine.

Check your own diary

Your diary can show whether chocolate came before the first signs of an attack or after them, and whether attacks are more common on the days you eat it. Here is a fictional diary kept for 8 weeks, or 56 days, with 10 attacks.

Fictional diary: days with and without chocolate over 8 weeks
DaysAttack beganShare
Chocolate eaten21419%
No chocolate35617%
All days561018%

In this fictional diary, chocolate was selected as a trigger for 4 of the 10 attacks, so a trigger summary would show it for 40% of attacks. That sounds high, but attacks began on 19% of chocolate days and 17% of other days, almost the same. In 3 of the 4 “chocolate” attacks, the notes record tiredness, yawning or a craving for sweets before the chocolate. With only 10 attacks, this proves nothing either way, but the notes fit chocolate being part of the build-up rather than a cause.

  1. Note the first sign, not just the pain. Write down when tiredness, yawning or a craving started, as well as when the headache began.
  2. Note when you ate chocolate. If the craving came first, write that down too.
  3. Count the days without an attack. A trigger list only covers attacks, so also keep a tally of the days you ate chocolate and had no attack, for example on a calendar.
  4. Compare shares, not counts. Divide attack days by all days, with and without chocolate, as in the table. The weekend migraines post uses the same check for Saturdays and Sundays, and the alcohol post uses it for evenings with a drink.
  5. Give it time. With a handful of attacks, a few percentage points can easily be chance. The guide to reading patterns explains why a pattern is a starting point, not proof of a cause.

What Migraine Tracker records

Migraine Tracker for iPhone and Apple Watch records suspected triggers, symptoms and notes with each attack, so you can see what you noted around an attack later.

  • Chocolate as a trigger. Migraine Tracker’s built-in food triggers include Chocolate, along with Caffeine Withdrawal, Excess Caffeine, Alcohol, Aged Cheese and others. You can select it during an attack or when you edit one. A trigger records what you suspected, not what caused the attack.
  • Notes. When you edit an attack or add a past one, you can write a note, for example when a craving began and when you ate.
  • Symptoms. Migraine Tracker’s built-in symptoms include Fatigue and Brain Fog, but no food craving. With Premium, you can add your own symptom, such as a craving.
  • Top Triggers. With Migraine Tracker Premium, Reports shows your 5 most common triggers in a period and the share of attacks that had each one. It does not count days without an attack, so it cannot tell a trigger from a habit; that is why the diary check above counts days.

The guide to what to record covers the other details worth keeping with each attack.

When to talk to someone

  • The NHS says to see a GP if “your migraine attacks are severe or getting worse, or lasting longer than usual”, if “you have migraine attacks more than once a week” or if “you’re finding it difficult to control your migraine”.
  • If you are thinking about cutting out chocolate or other foods, the Migraine Trust suggests getting advice from a GP or dietitian before significant diet changes. It advises against diets that cut out whole food groups without guidance from a healthcare professional.
  • The NHS says to call 999 if you “have a headache that came on suddenly and is extremely painful”. Elsewhere, call your local emergency number.

Information from the NHS website, as at 201026: Migraine.

Sources

Sources checked on 17 October 2026. The diary example uses fictional data.

App details checked against Migraine Tracker 1.0.9 on 17 October 2026. The diary example uses fictional data. This post explains research and record keeping, not medical or dietary advice. The sources explain general principles and do not endorse Migraine Tracker.