When to take migraine medication — the timing question

Short answer: most acute migraine medications work better when taken early — within 30 minutes of pain onset, while the headache is still mild. The PRODROME trial (2023) showed some abortives also work during the prodromal phase, before pain begins. The hard limit on the other end is medication overuse: taking acute medications too often can paradoxically worsen migraine over months.

This page lays out what published trials show. The specific decision of when and what to take is between you and your prescriber. Hermly forecasts the risk window; it does not prompt or recommend medication.

The “treat early” principle

For triptans, gepants, ditans, and most acute migraine medications, the consistent finding across decades of trials is that earlier treatment produces better response rates. The typical numbers:

• Treated within 30 minutes of pain onset (mild pain): pain-free at 2 hours in 50–70% of patients (varies by drug).
• Treated after pain becomes moderate-to-severe: pain-free at 2 hours in 30–45%.
• Treated after several hours of established pain: response drops further.

The proposed mechanism is central sensitisation — the trigeminal nervous system becomes hyper-responsive once the migraine cascade is fully underway, and abortive medications work better before that point. Allodynia (pain in response to normally non-painful stimuli — touch, light brushing) is the clinical sign that central sensitisation has occurred.

This is why guidance from headache specialists typically emphasises: take it at the first definite sign of attack, not “wait and see if it gets bad enough to need medication”.

The PRODROME trial: medication before pain begins

The 2023 PRODROME trial in Lancet tested the gepant ubrogepant (an oral CGRP receptor antagonist) taken during the prodromal phase — before pain onset — based on the patient’s own recognition of premonitory symptoms.

Result: significantly more patients in the treatment group avoided progression to moderate-or-severe headache in the following 24 hours, compared to placebo. This was a meaningful result — the first well-controlled evidence that a specific medication can be effective during the prodrome rather than waiting for pain.

The clinical implication: for patients who can reliably recognise their prodromal symptoms (see our prodrome guide) and who are prescribed an appropriate medication, treating during the prodrome may be more effective than waiting for pain to start.

Whether this applies to your specific medication and prescription is a conversation with your prescriber. Triptans during prodrome haven’t been tested as rigorously and behave differently from gepants.

The hard limit: medication overuse headache

Taking acute migraine medications too frequently can cause the condition to worsen over months. This is medication overuse headache (MOH), and it’s recognised in ICHD-3 as a distinct diagnosis.

The thresholds that define overuse vary by medication class:

• Triptans, ergots, opioids, combination analgesics: ≥ 10 days/month for 3+ months.
• Simple analgesics (ibuprofen, acetaminophen, aspirin): ≥ 15 days/month for 3+ months.

Crossing these thresholds risks transforming episodic migraine (few attacks/month) into chronic migraine (≥ 15 headache days/month) — exactly the trajectory most patients are trying to avoid.

This is why “take it earlier” is good advice up to a point, and not good advice if “earlier” means taking it more often. See our MOH learn page for the detail.

Situational prevention — a different approach

For predictable high-risk windows (menstrual migraine, planned travel, known stressful periods), some patients use situational prevention: a preventive medication taken only during the high-risk window, not daily.

Lipton et al. 2024 (Headache) outlined this approach formally, with examples like:

• Menstrual migraine — short-course triptans or NSAIDs taken for 5–6 days around menstruation.
• Exertional migraine — preventive dose taken before predictable exertion.
• Travel migraine — preventive on travel days for patients with reliable patterns.

This is prescription medicine, not over-the-counter. The choice of regimen and duration is specific to the patient and needs medical supervision.

What Hermly does and doesn’t do

Hermly forecasts risk. It does not:

• Prompt you to take medication.
• Recommend a specific medication or dose.
• Mark high-risk days with a “take your meds” badge.
• Auto-log medication intake.

What Hermly does:

• Surfaces today’s risk number with the three factors driving it.
• Lets you log medications taken (during attacks via Live Activity, or via Siri / Action Button intents).
• Tracks medication frequency over rolling 30-day windows so the data is available when you and your doctor look at the doctor report.

The reason for this restraint is twofold: regulatory (Hermly is not a medical device and cannot prescribe), and editorial (an app that nudges medication use on a forecast risks contributing to overuse for some users).

What to do with this information

The honest practical advice — much of which Hermly cannot give you itself, but which a clinician can:

• Have a prescribed plan before the next attack, not during it.
• Know your medication overuse limits and track them.
• Discuss situational prevention with your doctor if you have predictable high-risk windows.
• Consider preventive medication if attack frequency exceeds 4–6 days/month consistently.

Hermly’s job in all of this is to make the data legible — your attack frequency, your medication use, your risk windows — so you have something concrete to bring to that conversation.