Two Years of Cannabis for Migraines. The Pain Went Down. The Patients Aren’t Going Back.

A new longitudinal study tracked 203 refractory migraine patients on medical cannabis for up to two years. The results were consistent, sustained — and might change the conversation about access.

For patients who have tried everything — the triptans, the preventatives, the Botox injections every three months — refractory migraine isn’t just a headache. It’s a life organized around pain. Canceled plans. Lost workdays. The particular loneliness of a condition that’s invisible to everyone else and consuming to you.

A new study out of the United Kingdom is adding two years of hard data to what many of those patients already know firsthand: medical cannabis helps.

Published this week in Brain & Behavior, the research tracked 203 patients with a primary diagnosis of migraine who were enrolled in the UK Medical Cannabis Registry — a national database following patients who’ve been prescribed cannabis-based medicinal products (CBMPs) after failing conventional treatments. Researchers assessed outcomes at baseline and again at 1, 3, 6, 12, 18, and 24 months.

The results held across every interval.

Patients reported statistically significant improvements in migraine-related pain, disability, sleep quality, anxiety, depression, and overall health-related quality of life — and those gains didn’t erode over time. Two years in, the improvements were sustained.

What They Were Taking

Participants used either herbal cannabis or oil extracts containing standardized percentages of THC and CBD. The study didn’t prescribe a single formulation; instead it tracked what patients and their physicians chose within the registry, which is closer to how medical programs actually function in practice.

One finding stood out: patients consuming higher-THC products were most likely to report the greatest improvements. The study’s authors were careful to note that adverse events did occur, but characterized the majority as mild to moderate in severity.

The study’s conclusion pulled no punches: “The findings of this study show that CBMPs are associated with improved HRQoL [health-related quality of life] over 24 months in both headache-specific and other patient-reported outcome measures.”

Why This Study Is Different

Observational data has limits — the authors are clear on that, calling for randomized controlled trials to confirm what they’ve found. But a 24-month longitudinal study of this scale is genuinely uncommon in cannabis research, where funding constraints and federal prohibition in the U.S. have made long-term human studies difficult to conduct.

This is the third major finding to come out of the UK Medical Cannabis Registry in recent years covering distinct patient populations. Earlier reports from the same registry documented improvements in patients with treatment-resistant epilepsy, cancer-related pain, generalized anxiety disorder, inflammatory bowel disease, and endometriosis. The consistency across conditions — all observational, all using similar registry methodology — is starting to build a picture that is harder to dismiss.

For U.S. policymakers still debating whether cannabis belongs in clinical care, the registry model represents something America currently can’t build at scale: a long-running, multi-site, real-world data system for medical cannabis patients. The UK can do this in part because their regulatory framework treats prescribed CBMPs as legitimate medicine. Ours still doesn’t.

The Access Problem

In the United States, approximately 39 million people live with migraines. Of those, an estimated 15 to 20 percent are classified as refractory — meaning they don’t respond adequately to standard treatments. For that subset, options are limited and often expensive. Newer CGRP-targeting biologics can cost thousands of dollars a month before insurance intervention. Botox for migraine requires repeated clinic visits every 12 weeks and isn’t universally covered.

Cannabis, in states with medical programs, is often not covered by insurance at all — because federal prohibition means no Medicare or Medicaid reimbursement, and most private insurers won’t touch it. Patients in those states who want to try medical cannabis are largely paying out of pocket.

That’s the gap this research speaks to directly. Sustained, patient-reported improvement over two years isn’t a case study or an anecdote. It’s the kind of evidence that could — should — inform insurance coverage decisions, clinical guidelines, and state program design. Whether it does depends on whether policymakers and payers are willing to engage with data that doesn’t come wrapped in an FDA approval.

What Patients Already Know

The researchers’ call for randomized controlled trials is methodologically correct and practically complicated. Cannabis remains a Schedule I controlled substance under federal law in the United States, which means NIH-funded RCTs face bureaucratic friction that other drugs don’t. The DEA’s ongoing rescheduling process — still incomplete — may eventually ease some of those constraints. But that process has moved slowly, and the patients with refractory migraines aren’t waiting.

In medical cannabis states, many of them are already self-reporting what this study documented: reduced pain, better sleep, less anxiety, more functional days. The UK registry is notable because it puts that experience into a structured, longitudinal, peer-reviewed framework — which is exactly what advocates have been asking the U.S. to build for years.

Two years of data on 203 patients won’t settle every debate about medical cannabis and headache disorders. But it’s the most rigorous evidence yet that for people who have tried everything else, this isn’t a placebo effect. It’s a treatment.

Source: Observational data from the UK Medical Cannabis Registry, published in Brain & Behavior (Wiley), April 2026. DOI: 10.1002/brb3.71323. Study authors affiliated with British institutions; full author list available at the journal link.