Cannabis Beats Placebo for Migraines: First RCT Results

For two decades, migraine sufferers have been telling their neurologists that cannabis helps. For two decades, neurologists have been answering — accurately — that there were no randomized controlled trials behind the claim. That sentence is finally out of date.

A team led by Nathaniel M. Schuster, MD, at UC San Diego’s Center for Pain Medicine published the first randomized, double-blind, placebo-controlled trial of cannabis for acute migraine in Headache (Schuster et al., 2026; PMID 41469488). Vaporized flower containing 6% THC plus 11% CBD beat placebo on pain relief at 2 hours — 67.2% versus 46.6% (p=0.016) — and on every key secondary endpoint, with benefits sustained to 24 and 48 hours.

That is a real, placebo-controlled signal. It is also one trial, single-center, with about a hundred participants. This piece walks through what it showed, what it didn’t, and how to think about it.

Why This Trial Matters

Migraine affects roughly 1 in 7 adults and is the second-leading cause of years lived with disability worldwide. Treatment has improved — triptans in the 1990s, CGRP inhibitors since 2018 — but a third to half of patients still have inadequate relief, side-effect intolerance, or contraindications.

Into that gap stepped cannabis. By 2019, the Cuttler et al. Strainprint study analyzed nearly 20,000 self-tracked sessions and reported ~50% reductions in self-rated severity after inhaled cannabis — ecologically valid but unblinded. Until Schuster, we had observational signal and mechanism but no causal evidence. The new trial closes that loop. For consumer context on strains patients gravitate to, see our best cannabis strains for migraines and headaches guide.

What the Trial Did

This was a randomized, double-blind, placebo-controlled, crossover trial. The crossover design is important: every participant served as their own control, treating up to four separate migraine attacks — one with each of four conditions, in randomized order, with at least one week between attacks.

The four conditions were vaporized cannabis flower from the NIDA Drug Supply Program:

1. 6% THC (THC-dominant)
2. 11% CBD (CBD-dominant)
3. 6% THC + 11% CBD (the combination)
4. Placebo flower — cannabis with cannabinoids removed but matched on taste and smell

Ninety-two adults were enrolled; 247 attacks were treated. Mean age was 41, and 82.6% of participants were women — roughly representative of the migraine population.

Dosing: Four puffs from a precision vaporizer within the first four hours of a moderate-to-severe migraine attack. After the 2-hour primary assessment, rescue medication was permitted; no additional cannabis.

Primary endpoint: Pain relief at 2 hours per International Headache Society criteria — reduction from moderate or severe to mild or no pain.

Secondary endpoints: Pain freedom (no pain at all), freedom from the participant-selected most bothersome symptom (MBS — photophobia, phonophobia, or nausea), and sustained outcomes at 24 and 48 hours. Trial registration: NCT04360044. This IHS endpoint set is the same scaffolding used to evaluate triptans and CGRP inhibitors, making results directly comparable to the prescription migraine literature.

The Headline Result

Across 247 treated migraine attacks, the THC+CBD mix beat placebo on every primary and secondary endpoint at 2 hours:

Outcome at 2h	THC+CBD	Placebo	Odds Ratio (95% CI)	p-value
Pain relief	67.2%	46.6%	2.85 (1.22–6.65)	0.016
Pain freedom	34.5%	15.5%	3.30 (1.24–8.80)	0.017
MBS freedom	60.3%	34.5%	3.32 (1.45–7.64)	0.005

In plain English: out of every 100 attacks treated with THC+CBD, roughly 67 reached pain relief and 35 reached complete pain freedom at 2 hours. On placebo, those numbers were 47 and 16.

The p-values (all under 0.05) mean these differences are highly unlikely to be chance — p=0.005 for MBS freedom is about a 1-in-200 probability of a result this large from random variation alone. The odds ratios of 2.85 to 3.32 say participants on THC+CBD were roughly three times as likely to reach each outcome as on placebo. Results held across modified intent-to-treat and per-protocol analyses.

What the Single-Compound Arms Showed

This is where the science gets interesting:

• THC-dominant (6% THC alone) beat placebo for pain relief (68.9% vs 46.6%, p=0.008) but not for pain freedom or MBS freedom at 2 hours.
• CBD-dominant (11% CBD alone) was not superior to placebo on any of the three primary or secondary endpoints at 2 hours.

Pure CBD flower didn’t move the needle on acute migraine in this trial. THC alone reduced pain but didn’t reliably eliminate it or knock out the most bothersome symptom. The combination drove the strongest, most complete results.

That’s meaningful for the consumer market, where CBD-only products are often marketed for headache without strong evidence. This trial suggests CBD’s contribution shows up in combination with THC, not as a solo acute treatment. Whether daily preventive CBD has effects is a separate question.

The combination was also better tolerated than THC alone — lower rates of euphoria, cognitive impairment, and reported “highness” — consistent with CBD blunting some of THC’s psychoactive intensity without eliminating analgesia.

Sustained Outcomes — and Why They Matter

For migraine patients, the question after “did the pain go away?” is “did it stay gone?” Triptans sometimes work fast and then the migraine recurs. The THC+CBD arm beat placebo on 24-hour sustained pain freedom, 24- and 48-hour sustained MBS freedom, and 2-hour photophobia and phonophobia freedom.

Sustained outcomes separate genuine acute treatments from temporary masking. The THC+CBD benefit holding to 48 hours is one of the paper’s more clinically meaningful findings.

What This Trial Doesn’t Prove

Honest science means naming the limits:

1. Single-center, modest N. 92 enrolled, 247 attacks analyzed — enough for primary-endpoint power, not enough to detect rare subgroup effects or replicate across populations. Multicenter replication is the explicit next step.

2. No active comparator. Cannabis was compared to placebo, not to triptans, CGRP inhibitors, or NSAIDs. We don’t know whether THC+CBD flower works as well as, better than, or worse than sumatriptan — the comparison patients actually need.

3. Blinding is hard with cannabis. Real THC is felt. The authors’ blinding analysis was reassuring but not perfect; some unblinding effect on subjective endpoints is plausible.

4. Substantial placebo response. 47% of placebo attacks reached pain relief — typical of acute migraine trials, but a reminder that the cannabis benefit is the additional effect on top of placebo, not the entire effect.

5. Short-term protocol. Four puffs, single attack, week-long washouts. The trial says nothing about daily preventive use, long-term safety, medication-overuse headache risk, or cannabis use disorder incidence.

6. NIDA flower isn’t dispensary flower. Federal research-grade cannabis differs in terpenes and potency from commercial product. The 6% THC tested is well below typical dispensary flower (15–30%), so results may not translate cleanly to high-THC product.

The Mechanism: Why Might It Work?

The biology has been suggestive for decades. Russo’s clinical endocannabinoid deficiency theory (2004; reconsidered 2016) proposed migraine, fibromyalgia, and IBS may share an underlying deficit in endocannabinoid tone. Key evidence: CSF anandamide is measurably lower in migraineurs; CB1 receptors sit on the trigeminovascular fibers that convey migraine pain; the periaqueductal gray (a migraine generator) is rich in endocannabinoid signaling; anandamide potentiates 5-HT1A and inhibits 5-HT2A — the same serotonin targets triptans hit; and CGRP, the molecule modern migraine drugs target, is modulated by endocannabinoid signaling.

The two-cannabinoid synergy makes biological sense. THC binds CB1 directly; CBD acts indirectly — modulating CB1 allosterically, inhibiting FAAH (raising anandamide), and engaging TRPV1 and 5-HT1A. The combination plausibly hits multiple migraine-relevant pathways at once.

This is why our Relief family emphasizes balanced profiles — caryophyllene at CB2 receptors, myrcene for muscle relaxation, linalool for calming analgesia, and pinene for anti-inflammatory effects map onto pain pathways the Schuster trial only began to probe.

Where This Fits in the Broader Literature

The Schuster RCT corroborates a substantial pre-existing observational signal:

• Cuttler et al. (2020) — Strainprint analysis of 12,293 headache and 7,441 migraine sessions: ~47–50% reductions in self-rated severity, with tolerance developing over time.
• Stith et al. (2019, 2020) — Releaf app data: 94% of users reported symptom relief within two hours, with flower more effective than concentrate.
• Pini et al. (2012) — Small open-label trial of nabilone (synthetic THC) showing benefit over ibuprofen for medication-overuse headache.

The Schuster trial converts that signal into causal evidence — the difference between “patients say it works” and “we have a placebo-controlled demonstration.” Both can be true at once.

For broader cannabinoid analgesia, see our cannabis vs. opioids deep-dive and arthritis and joint pain guide.

What This Means For You — Practically

Four points of practical information:

1. Product type: Vaporized whole flower. Not edibles, tinctures, concentrates, or isolates.

2. Cannabinoid ratio: Roughly 1:2 THC to CBD. 6% THC + 11% CBD is close to a 1:2 ratio favoring CBD — dispensary terms are “balanced” or “high-CBD with THC.” The Balance family catalogs strains in this neighborhood.

3. Dose: Four measured puffs from a vaporizer, within the first four hours of a migraine attack.

4. Single-cannabinoid arms didn’t work as well. Pure CBD flower failed every endpoint. Pure THC flower hit pain relief but missed pain freedom and MBS freedom. The combination is what mattered.

If you want to map this onto dispensary cultivars, balanced or CBD-leaning options include Harlequin, Cannatonic, ACDC, and moderate-THC classics like Blue Dream. Heavy-THC cultivars like OG Kush or Sour Diesel — popular in real-world migraine self-medication — were not the configuration tested.

This is not medical advice and is not a replacement for your neurologist’s care. Triptans and CGRP inhibitors are well-evidenced and often life-changing. Cannabis is now an evidence-supported addition to consider, not a substitute. If you’re already on migraine medications, talk to your physician before adding cannabis — there are documented interactions, particularly with drugs metabolized by CYP3A4 and CYP2C9.

The Personalization Problem

Even a positive RCT can’t tell you what will work for your migraines. Migraine is heterogeneous — some attacks respond to triptans, others don’t; some need NSAIDs, others a dark room. Cannabis won’t be different. Inside the trial averages were people who got dramatic relief and people who got nothing. The only way to know which group you’re in is to track carefully.

Cannabis is also a multi-compound matrix. The same cultivar from two farms can have different terpene profiles, and two products labeled “1:2 THC:CBD” can feel meaningfully different. Set, setting, hydration, sleep, and where you are in the attack timeline all shift outcomes.

So: track what works, log what doesn’t, and let your own data accumulate. Twenty migraines logged systematically — cultivar, ratio, route, dose, time-since-onset, outcome — start to tell a real story. That’s the kind of pain relief and headache relief data only your body can generate.

What Comes Next

The Schuster trial is a beginning. Realistic next steps: multicenter replication with larger, more diverse populations; active-comparator trials versus triptans and CGRP antagonists; longer follow-up for medication-overuse headache and cannabis use disorder risk; dose-finding (is four puffs optimal?); preventive trials of daily low-dose cannabinoid therapy; and subgroup analyses across migraine subtypes and triptan non-responders. Pharmaceutical-grade formulations like nabiximols may eventually be tested here, addressing the dosing-precision concerns inherent in flower.

Sources

Primary trial:

• Schuster NM, Wallace MS, Marcotte TD, Buse DC, Lee E, Liu L, Sexton M. Vaporized cannabis versus placebo for acute migraine: A randomized, double-blind, placebo-controlled crossover trial. Headache. 2026;66(2):365-376. doi: 10.1111/head.70025. PMID: 41469488. PubMed | PMC | NCT04360044

Supporting literature:

• Cuttler C, et al. Short- and Long-Term Effects of Cannabis on Headache and Migraine. J Pain. 2020;21(5-6):722-730. PubMed
• Russo EB. Clinical Endocannabinoid Deficiency Reconsidered. Cannabis Cannabinoid Res. 2016;1(1):154-165. Liebert
• Russo EB. Clinical endocannabinoid deficiency (CECD). Neuro Endocrinol Lett. 2004;25(1-2):31-39. PubMed
• Schuster NM et al. Vaporized Cannabis versus Placebo for Acute Migraine (S22.010). Neurology. 2024;102(17 Suppl 1). Neurology.org

A Word From Professor High

For decades, patients have known something the formal evidence base couldn’t yet confirm. The Schuster trial is the field catching up. The next decade of work is what will turn “promising” into standard of care, if it gets there.

Track what works for you. Bring data, not anecdotes, to your neurologist. If you try cannabis for migraine, the trial’s specifics — vaporized flower, balanced THC and CBD, low-to-moderate dose, taken early — are a far better starting point than improvising. Open your TIWIH log here and start building your own data.

---

Medical Disclaimer

This article is educational and is not medical advice. Migraine is a serious neurological condition that warrants evaluation and treatment by a qualified healthcare provider. Do not start, stop, or substitute migraine medication based on this article. Cannabis can interact with prescription medications, may not be appropriate during pregnancy or with certain medical or psychiatric conditions, and carries risks including impairment, dependence, and — with frequent use — medication-overuse headache. If you are considering cannabis for migraine, discuss it with your neurologist or primary care physician first.