Headache is one of the most common reasons people give for using cannabis as medicine, and one of the least studied. This article covers four things. It explains the idea behind using cannabis for headache and the one controlled trial for migraine attacks. It describes the overuse trap that can catch frequent users. It also sets out what is still unknown.
This article is educational. It does not recommend treating migraine with cannabis. If you have migraine or cluster headache, see a neurologist (a doctor who specialises in the brain and nerves).
Translator’s note: Growers in Thailand and Southeast Asia are often asked for “a strain for migraine”. This article helps you answer honestly. It also shows why a useful answer depends on tested cannabinoid content, not on a strain name.
The idea behind it: clinical endocannabinoid deficiency
Your body makes its own cannabis-like signalling molecules, called endocannabinoids. They act on the same receptors as THC. Together with those receptors they form the endocannabinoid system.
The argument for cannabis in headache comes from an idea called clinical endocannabinoid deficiency. The researcher Ethan Russo proposed it and looked at it again in a 2016 review in the journal Cannabis and Cannabinoid Research. It suggests that some hard-to-treat conditions, including migraine, share a low background level of endocannabinoid signalling (low “endocannabinoid tone”).
Supporters point out that CB1 receptors, the main receptors THC binds to, are found in the brain structures involved in migraine. But the idea is still a hypothesis, not a proven mechanism. It is a good reason to run trials. It is not a reason to expect those trials to succeed.
What you can do: if someone says “endocannabinoid deficiency” proves that cannabis treats migraine, tell them it is an untested idea. Then show them the trial results below.
The one controlled trial for migraine attacks
Almost everything else published on cannabis and headache is survey data, where people report their own experience. There is now one properly designed randomised controlled trial for acute migraine, meaning treatment of an attack after it has started. Because it is the only one, it is worth knowing in detail.
Who ran it and how
Nathaniel Schuster, a researcher at the University of California San Diego (UC San Diego, United States), led the trial. It was published in the journal Headache in 2026, after appearing online in December 2025.
The trial design had four features:
- Randomised: chance decided which treatment each attack received.
- Double-blind: neither the participants nor the researchers knew which treatment was used each time.
- Placebo-controlled: one treatment was a dummy flower with no THC or CBD. This shows how much people improve without any active ingredient.
- Crossover: each participant used different treatments on different attacks, so each person is compared with themselves.
Each participant treated up to four separate migraine attacks, at least one week apart. Each attack was treated with vaporised flower from a different, randomly assigned group. A vaporiser heats flower until it releases vapour, without burning it. 92 participants were enrolled and randomised, and 247 migraine attacks were treated.
| Treatment group | Content of the flower |
|---|---|
| THC-dominant | 6% THC |
| CBD-dominant | 11% CBD |
| THC + CBD | 6% THC + 11% CBD |
| Placebo | no THC or CBD |
How to read the numbers
The tables below use three statistical terms. You only need their basic meaning:
- OR (odds ratio): how much a treatment raised the odds of a good result compared with placebo. 1 means no difference. About 3 means roughly three times the odds.
- Range in square brackets (confidence interval): the range where the true odds ratio probably lies. If the whole range is above 1, the benefit is unlikely to be a fluke.
- p value: the chance of seeing a difference this large if the treatment did nothing. Below 0.05 is the usual cut-off for “statistically significant”.
Main result: pain relief after 2 hours
The trial’s primary endpoint, the main result it was designed to measure, was pain relief 2 hours after treatment. Pain relief meant pain falling from moderate or severe to mild or none.
| Group | Pain relief at 2 hours | Compared with placebo |
|---|---|---|
| THC-dominant | 68.9% | OR 3.14 [1.35–7.30], p = 0.008 |
| THC + CBD | 67.2% | OR 2.85 [1.22–6.65], p = 0.016 |
| CBD-dominant | — | not better than placebo |
| Placebo | 46.6% | — |
In plain words, both groups with THC gave pain relief about two times in three. Placebo gave relief about one time in two. CBD alone did no better than placebo.
Pain freedom and freedom from the worst symptom
The trial also measured secondary endpoints, which are extra results recorded alongside the main one. Two of them clearly separated the groups. “Pain freedom” means no pain at all after 2 hours. “Most bothersome symptom (MBS) freedom” means the worst symptom other than pain had gone. Before treating each attack, participants chose their most bothersome symptom from three: sensitivity to light, sensitivity to sound or nausea.
| Result at 2 hours | THC + CBD | Placebo | Compared with placebo |
|---|---|---|---|
| Pain freedom | 34.5% | 15.5% | OR 3.30 [1.24–8.80], p = 0.017 |
| MBS freedom | 60.3% | 34.5% | OR 3.32 [1.45–7.64], p = 0.005 |
The THC-dominant group did not beat placebo on either of these results. The CBD-dominant group did not beat placebo on any result. Only the combination did.
Lasting benefit and side effects
Only the combination gave lasting benefit. The THC + CBD group beat placebo for pain freedom that still held at 24 hours. It also beat placebo for MBS freedom that still held at 24 and 48 hours.
No serious adverse events (serious harmful effects) occurred in any group. The investigators reported that THC side effects were milder with THC + CBD than with THC alone.
What the results mean
For pain relief at 2 hours, THC alone and the combination performed about the same: roughly 68%, against 47% for placebo. Only the combination also gave pain freedom, freedom from the worst symptom and benefit lasting 24 to 48 hours. It also seems to have caused less intoxication than THC alone. So in this trial, adding CBD did not weaken the benefit.
The placebo response is large
Placebo flower gave 46.6% pain relief and 15.5% pain freedom. A placebo response this large is normal in migraine trials. It is why personal reports that cannabis “works” for headache tell you so little: many of those headaches would have eased without any active ingredient.
Limits of the trial
- It tested only THC and CBD. It did not test minor cannabinoids or terpenes (the aroma compounds).
- It used one strength of each flower, one THC:CBD ratio and only inhaled flower.
- It did not study repeated or long-term use.
So the trial tells you nothing about three things: the risk of medication overuse headache (see below), cannabis use disorder (use that a person finds hard to control) and migraine prevention.
What you can do: if you discuss this trial with someone, state its exact conditions. It tested vaporised flower, 6% THC with 11% CBD, one attack at a time. Do not stretch the result to oils, edibles, daily use or other ratios.
Prevention and cluster headache: no controlled trials
There are no randomised controlled trials of cannabis for preventing migraine. Treating an attack and preventing attacks are separate questions, and each needs its own evidence. Only the first one has been tested.
Cluster headache is a separate headache disorder with very severe attacks. It also has no randomised controlled trials of cannabis, and the survey data is not encouraging. A French team led by Leroux published a study of 139 cluster headache patients in the headache journal Cephalalgia in 2013. Of these patients, 27 had tried cannabis for attacks:
- About a quarter reported some benefit.
- About half reported variable or uncertain effects.
- About a fifth said it made attacks worse.
The authors concluded that cannabis has limited effect on cluster attacks. Anyone who says cannabis is proven for cluster headache is going beyond the evidence.
What you can do: if someone asks about cannabis for migraine prevention or cluster headache, tell them plainly that no controlled trial exists. Refer them to a neurologist.
Medication overuse headache: the trap for frequent users
This part of the subject gets the least attention, even though it matters most.
What it is
Medication overuse headache is a recognised medical condition. If you take headache medicine on too many days, occasional headaches can turn into headaches on almost every day. The treatment becomes the cause.
It is known to happen with these medicines:
- triptans, prescription medicines taken at the start of a migraine attack
- combination painkillers, tablets with two or more pain ingredients
- opioids, strong painkillers such as codeine
- simple painkillers, once they are taken often enough
Patients usually do not notice the problem. It feels as if the migraine is getting worse, so they take more medicine.
Does cannabis cause it?
Nobody knows yet. A review titled “Cannabinoids and Psychedelics for Headache Management” appeared in the medical journal Practical Neurology in May–June 2024. It says: “No RCTs have been performed to study cannabis for migraine prevention, and additional studies are needed to evaluate whether cannabis use contributes to medication overuse headache in some individuals.” (RCT means randomised controlled trial.)
So it is plausible that cannabis can cause medication overuse headache. No study has shown whether it does. Nothing known so far rules it out.
The official thresholds
You can protect yourself without waiting for an answer: count how often you treat headaches. Doctors diagnose headache types using the International Classification of Headache Disorders, 3rd edition (ICHD-3). The International Headache Society publishes it. Under ICHD-3, medication overuse headache requires both of these:
- headache on 15 or more days a month in someone who already had a headache disorder
- regular overuse of attack medicine for more than three months
| Type of medicine | Counts as overuse at |
|---|---|
| Triptans or combination painkillers | 10 or more days a month |
| Simple painkillers: paracetamol, or NSAIDs (anti-inflammatory painkillers such as ibuprofen) | 15 or more days a month |
| Any other medicine taken to treat headache (cannabis is not named) | 10 or more days a month |
If you treat headaches with anything, cannabis included, on that many days, discuss it with a neurologist.
How to keep a headache diary
A headache diary is the most useful thing a patient in this situation can produce, and it costs nothing. It records the days with headache and the days with treatment. A paper notebook or a phone note is enough.
Write one line each day:
- The date.
- Headache yes or no. If yes, write how bad it was (0–10) and how long it lasted.
- Anything you took for it: the name, the amount and the time. Include cannabis, with how much you used and how you took it.
- The effect after 2 hours: no change, better, or gone.
At the end of each month, count two numbers: days with headache and days with any treatment. Compare them with the table above. Take the diary to your neurologist, because it shows patterns that memory misses.
Cannabinoid hyperalgesia
Cannabinoid hyperalgesia is a related effect. In some situations, heavy long-term cannabinoid use can make a person more sensitive to pain, not less. The general pain literature describes it. Nobody has shown yet whether it matters for headache.
What you can do: start a headache diary this month, before you change anything. After 30 days you will know your real number of headache days and treatment days.
Older advice, strain names and what growers can test
Much of the advice written before this trial was reasonable in parts. It said the evidence came mostly from animal studies. It said some patients got worse, and that people should start with a low dose. But it often sent readers to strain-recommendation websites, which collect user reports and are not clinical evidence.
The trial suggests that what matters is the cannabinoid content and the THC:CBD ratio, not the cultivar (strain) name. One strain name can carry very different THC and CBD levels from one grower to the next. The chemotype system groups plants by their actual THC:CBD balance.
Translator’s note: For growers, this means a strain name tells a patient very little, while a lab result tells them a lot. If you grow under licence for medical supply, have every batch tested. Keep the certificate of analysis (the lab report of THC and CBD percentages) linked to the batch number. A prescriber can then see the real content and does not have to guess from a name.
Translator’s note: Rooms in Thailand often reach 30–35 °C, and heat, light and air break cannabinoids down over time. A test done at harvest may no longer match flower stored for months in a hot room. The guide to storing dried flower explains how to slow this down. What you can do: store tested batches somewhere cool and dark, and write the test date on the label.
Where this leaves a patient
- Migraine attacks: there is now one randomised controlled trial. It showed a benefit from vaporised THC flower. THC combined with CBD gave a more consistent benefit across the results, with less intoxication. That is real evidence, but it is only one trial.
- Migraine prevention: there is no controlled evidence.
- Cluster headache: there is no controlled evidence. Survey results are mixed or negative.
- How often you use it: the overuse risk is proven for similar attack medicines but not yet tested for cannabis. Watch how often you use it.
- THC with CBD: one trial with fixed strengths of inhaled flower suggests that adding CBD to THC may reduce intoxication without losing benefit for attacks. That is not enough to recommend any product. Discuss it with your doctor.
This is not a migraine treatment recommendation. Anyone with migraine or cluster headache should be assessed by a neurologist.
Translator’s note: When you see a neurologist in Thailand, take your headache diary. Tell the doctor about any cannabis use and how much you use. Cannabis can interact with other medicines, so bring the drug interaction list as well.
Thai rules (as of October 2026)
Translator’s note: This section covers the Thai rules that apply to anyone hoping to use cannabis flower for headache, and to growers who are asked to supply it.
Cannabis flower (ช่อดอกกัญชา) is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis), B.E. 2568. The notification was issued under the Thai Traditional Medicine Protection Act B.E. 2542 (1999). It has been in force since 26 June 2025. Its key points:
- Flower may be sold only against a prescription on form ภ.ท. 33 (PT 33) from an authorised practitioner.
- Each prescription covers no more than 30 days.
- Flower that licensees sell or export must come from a source with GACP certification from the Department of Thai Traditional and Alternative Medicine (GACP means Good Agricultural and Collection Practice).
- Advertising and online sale of flower are prohibited. Selling flower for smoking at the business premises is also banned, except when a licensed practitioner treats their own patients.
The notification and related rules decide which professions count as authorised practitioners. Check the current list with the Department of Thai Traditional and Alternative Medicine, not with lists shared online. Reports also differ on whether the notification names specific conditions. Ask your prescribing practitioner.
Translator’s note: This article cannot confirm that licensed shops stock flower with a ratio like the trial’s (6% THC with 11% CBD). If a prescriber suggests flower, ask the shop for that batch’s certificate of analysis. It shows the real THC and CBD content.
Translator’s note for growers: selling flower without a licence to a friend or customer with migraine is a criminal offence. Under Section 78 of the 1999 Act, unlicensed sale of a controlled herb (Section 46) carries up to 1 year in prison, a fine of up to 20,000 baht, or both. What you can do: refer anyone who asks for “migraine flower” to a neurologist and to an authorised practitioner who can issue a PT 33.
2026 updates. In 2026, two Ministerial Regulations of B.E. 2569 tightened licensing (see the two 2026 regulations):
- Cannabis or Hemp Extracts: in force since 26 April 2026. Extracts above 0.2% THC may be licensed only to three kinds of holder: Thai juristic persons that are not “foreign” under the Foreign Business Act B.E. 2542, state agencies, and the Thai Red Cross Society.
- Controlled Herbs (No. 2): in force since 30 April 2026. It tightens the licence criteria for cannabis flower.
In June 2026 the authorities stepped up enforcement of PT 33 prescriptions (see the June 2026 crackdown). The draft Cannabis Control Act is not law yet. The Cabinet approved the draft on 22 September 2026 and sent it to the House. For the full picture, see Thailand’s cannabis rules, 2022 to 2026.
What you can do: before you act on any of this, read the original legal text and check the date of the latest rule.
Questions and discussion
If you have kept a headache diary, or have a question about the trial or the Thai rules, post it below this topic. Questions about your own dose belong with your doctor.
References
- Schuster NM, Wallace MS, Marcotte TD, et al. “Vaporized cannabis versus placebo for acute migraine: A randomized, double-blind, placebo-controlled crossover trial.” Headache: The Journal of Head and Face Pain, 2026;66(2):365–376 (published online 30 December 2025). doi:10.1111/head.70025. Earlier preprint: medRxiv 2024, doi:10.1101/2024.02.16.24302843.
- Leroux E, et al. “Use of cannabis among 139 cluster headache sufferers.” Cephalalgia, 2013. doi:10.1177/0333102412468669.
- Headache Classification Committee of the International Headache Society. ICHD-3, section 8.2: Medication-overuse headache.
- “Cannabinoids and Psychedelics for Headache Management.” Practical Neurology, May–June 2024.
- Russo E. “Clinical Endocannabinoid Deficiency Reconsidered.” Cannabis and Cannabinoid Research, 2016.
- Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis), B.E. 2568.