Migraine triggers, keeping a diary that is actually useful

Almost everyone with migraine is told to keep a trigger diary. Most of those diaries produce either nothing or a wrong answer, and the reasons are specific and fixable.

The commonest outcome is a list of foods that are not really triggers, and an unnecessarily restricted diet. That is worse than having no diary.

This article is about how to keep one that works, what to record, what the statistical trap is, and how to actually test a suspicion.

This is information, not medical advice. A sudden, severe, or changing headache needs assessment before any diary.

Why most trigger diaries fail

Four reasons, and all four are avoidable.

1. The prodrome problem, and this is the big one

Migraine attacks begin before the pain. There is a premonitory phase (prodrome) lasting hours to a day or more, with yawning, mood change, neck stiffness, fatigue, food cravings, frequent urination and light sensitivity.

And here is the trap: the prodrome itself can cause the behaviour you then blame.

A craving for chocolate is a recognised prodromal symptom. So someone eats chocolate, the headache arrives, and they conclude chocolate caused it.

In reality the attack had already started, and the craving was one of its symptoms. Cause and effect are the wrong way round.

This has been tested, and it is why chocolate’s status as a trigger is genuinely doubtful, despite being the most famous food trigger in the world.

The same applies to: wanting a strong coffee or tea (prodromal fatigue), wanting something sweet, a bright room suddenly feeling unbearable (prodromal photophobia), and a stiff neck that you then blame on your posture.

So the first rule of a useful diary: record the prodromal symptoms as well as the headache. If a “trigger” keeps appearing alongside yawning and neck stiffness, you are watching the attack, not its cause.

2. The statistical trap

If you have, say, five headaches a month, and you also eat ten different foods a day, then some food will coincide with every headache by chance.

And the brain is very good at spotting coincidences and very poor at noticing the non-events, the days you ate the same food and had no headache.

So the second rule: record the headache-free days too, with the same detail. A diary that only records attack days cannot distinguish a trigger from a coincidence. This is the single most common design flaw.

3. The lag

A trigger does not usually produce a headache within minutes. The typical interval is several hours up to about 24, sometimes longer.

So people look at the wrong window, either at what they ate just before the pain, or at the whole of the previous week.

Third rule: record events with times, so you can look at the 6–24 hour window retrospectively rather than guessing.

4. Triggers usually need to stack

And this is why “it caused it last time but not this time” is so confusing.

Most migraine does not have single, reliable triggers. It has a threshold, and several smaller factors combine to cross it.

A poor night’s sleep alone may be fine. A poor night plus a skipped lunch plus the day before a period may not be.

Fourth rule: do not look for the trigger. Look for the combinations that cross your threshold. This changes what you do with the information, you manage the stack rather than eliminate one food.

What to record

Keep it short enough to actually do. A diary nobody fills in is worthless, and over-detailed diaries get abandoned in a fortnight.

Every day, including good days:

• Date
• Headache: yes/no. If yes: start time, end time, severity 1–10, side, and whether it stopped you doing something
• Prodromal symptoms, yawning, neck stiffness, mood change, cravings, tiredness, light sensitivity
• Aura, yes/no, and what
• Medicine taken, what, how many, and what time. This is the most important single column
• Sleep, hours, and wake time
• Meals, whether any were skipped or late (not every item)
• Period, and where you are in the cycle
• Stress, a 1–5 number is enough
• Exercise, yes/no
• And anything unusual, travel, a late night, alcohol, a new medicine, illness

What to leave out at first:

A full food log. Start with meal timing rather than meal contents, because skipped and delayed meals are far better-evidenced triggers than specific foods, and a full food log is both laborious and the main source of false conclusions.

Only add specific foods later, if the pattern suggests it.

And note: the medicine column is the one that most often produces an actionable finding, because it reveals medication-overuse headache, which is common and treatable, and which no amount of food analysis will show.

How long, and how to read it

Three months. Less than that in a condition with monthly hormonal variation and month-to-month fluctuation will mislead you.

And then look for, in this order:

1. Days per month with headache, and days per month with medicine. Over 10–15 medicine days a month is the finding that matters most, and it changes the plan entirely.

2. Whether attacks cluster around the period. A clear pattern means menstrual migraine, which is planned for differently.

3. Sleep. Both short sleep and long sleep. Weekend lie-ins showing up before attacks is a very common and very fixable finding.

4. Skipped or late meals.

5. Then, and only then, specific foods, and test them properly, below.

And take it to your doctor. A three-month diary is genuinely more useful to them than any description from memory, and it is what the decision about preventive treatment is made on.

The triggers with reasonable evidence

Being clear about which are well supported, because the popular list and the evidenced list are different.

Reasonably well supported:

Skipped or delayed meals, and fasting. Sleep, too little and too much, and irregular timing. Menstruation, specifically the drop in oestrogen. Stress, and the let-down after stress, the Friday-evening or first-day-of-holiday attack is a recognised pattern. Alcohol, particularly red wine for some people. Dehydration. Caffeine withdrawal, and this is why the weekend attack is so common. Weather changes, patients report it consistently; studies are mixed, which is an honest position rather than a dismissal. Bright or flickering light. Strong smells. And exertion, particularly sudden and unaccustomed.

Much weaker than their reputation:

Chocolate, see the prodrome problem. Cheese and tyramine, the evidence is weaker than the folklore. MSG, poorly supported in controlled studies, despite being the most-cited food additive. Aspartame, weak. Nitrates in processed meat, plausible, limited evidence.

And this is not to say your own trigger is imaginary. Individual triggers exist and are real. It is to say that the standard “migraine food list” circulating online has far less behind it than people assume, and following it wholesale costs you a varied diet for little return.

How to test a suspected trigger properly

This is the part that is almost always skipped.

If your diary suggests something, test it deliberately:

1. Keep everything else stable, sleep, meals, caffeine, for the test period.

2. Expose yourself to the suspected trigger on three separate occasions, ideally when you are otherwise low-risk (good sleep, not premenstrual, not stressed).

3. Record what happens in the following 24 hours.

4. And then avoid it for a month and see whether attack frequency changes.

If three exposures in low-risk conditions produce nothing, it is probably not your trigger, or it is only a trigger when stacked with others, which is a different and more useful conclusion.

And the reason this matters: unnecessary food restriction is a real harm. It narrows the diet, creates anxiety about eating, makes social life harder, and in people who then skip meals to avoid “trigger foods” it can make migraine worse, because skipped meals are a better-evidenced trigger than most of the foods being avoided.

The apps, briefly

Several migraine diary apps exist and they are a reasonable choice, they prompt you daily, which solves the main compliance problem, and they generate summaries.

And a paper diary works perfectly well. The format matters much less than including the good days and the medicine count.

One caution: if you use an app, check what happens to your data before entering health information.

What to do with the findings

Manage the stack, not the list.

The things with the best return:

A fixed wake time, seven days a week. This single change addresses short sleep, long sleep, weekend lie-ins and caffeine-withdrawal timing all at once, which is why it is the highest-value intervention in this article.

Regular meals. Do not skip lunch.

Consistent caffeine, or none, the problem is variation.

Hydration.

Regular, moderate exercise, build up gradually.

And for the stack you cannot control, the period, the weather, a stressful week, know that those are the weeks to be stricter about the ones you can.

And if you are having more than about four attack days a month, or attacks that stop you working, ask about preventive treatment. Effective preventives exist, they take 8–12 weeks to show their effect, and most people are never offered them, which is the biggest gap in migraine care in India.

Claims, our position

What is sayable: riboflavin and magnesium contribute to normal energy-yielding metabolism, normal functioning of the nervous system and the reduction of tiredness and fatigue; magnesium contributes to normal muscle function and normal psychological function; and traditional-use statements for nirgundi and jatamansi.

What is not sayable: prevents migraine, reduces attack frequency, treats headache, or identifies or neutralises triggers.

And the honest point of this article: a diary costs nothing and is more likely to change your outcome than anything we sell.

Its two most valuable outputs are a medicine count that reveals medication-overuse headache, and a sleep pattern that reveals a fixable one, and neither involves a product.

When to see a doctor

Urgently, and do not wait to fill in a diary:

• A sudden, severe headache, the worst of your life
• Headache with fever and a stiff neck
• Headache with weakness, numbness, speech difficulty or visual loss
• Headache after head injury
• A new headache over 50, or in pregnancy, or with cancer or immunosuppression
• Headache worse on coughing, straining or lying flat
• Seizure or confusion

Soon:

• More than 10 medicine days a month, medication-overuse headache
• More than about 4 attack days a month, ask about preventive treatment
• A change in your usual pattern
• Attacks clearly tied to your period, tell your doctor
• If you have aura and take an oestrogen-containing contraceptive, this needs discussing; stroke risk is raised
• Morning headaches with snoring, sleep apnoea
• Blood pressure, vision and haemoglobin not checked

The short version

1. Most trigger diaries fail for four specific, fixable reasons.
2. The prodrome problem is the biggest: attacks begin before the pain, and the prodrome causes the behaviour you then blame.
3. A chocolate craving is a recognised prodromal symptom, which is why chocolate’s status as a trigger is doubtful.
4. Record prodromal symptoms, yawning, neck stiffness, cravings, mood.
5. Record the headache-free days too. A diary of attack days alone cannot separate trigger from coincidence.
6. Record times, the lag is typically 6 to 24 hours.
7. Triggers stack. There is a threshold, not a single cause.
8. Manage the stack, not the list.
9. Keep it short, a diary you abandon is worthless.
10. The medicine column is the most important one.
11. Start with meal timing, not a full food log, skipped meals are better evidenced than most foods.
12. Three months minimum, because of monthly variation.
13. Read it in order: medicine days, cycle, sleep, meals, then foods.
14. Over 10–15 medicine days a month changes the whole plan.
15. Weekend lie-ins before attacks is a common and very fixable finding.
16. Well-supported triggers: skipped meals, sleep change, menstruation, stress let-down, alcohol, dehydration, caffeine withdrawal, light, smells, exertion.
17. Weaker than their reputation: chocolate, cheese, MSG, aspartame.
18. Your individual trigger may still be real, the generic online list is what is weak.
19. Test a suspicion three times, in low-risk conditions, then avoid for a month.
20. Unnecessary food restriction is a real harm, and skipping meals to avoid “trigger foods” makes migraine worse.
21. A fixed wake time seven days a week fixes four problems at once.
22. Consistent caffeine, or none, variation is the problem.
23. More than 4 attack days a month: ask about preventives. They take 8–12 weeks.
24. Most people are never offered preventive treatment, the biggest gap in migraine care here.

Composition detail: Migro 360. Related: सिर दर्द के कारण और उपाय, riboflavin and magnesium, the prevention evidence, and माइग्रेन में क्या नहीं खाना चाहिए.