What It Treats & The Evidence
This is the honest heart of it: strong evidence for a few specific things, and much thinner evidence for the popular ones. Here's the difference, graded.
Ask what medical cannabis treats and you'll get two very different answers depending on who you ask. The marketing version is "almost everything". The evidence version is narrower and more honest: it's genuinely well-proven for a handful of specific conditions, modestly useful for a couple more, and for many of the things people most want it for, the evidence is still thin. That gap is the single most important thing on this page, so we've graded every claim by how strong the research behind it really is. Tap any badge to see the source.
Where the evidence is strongest: rare epilepsies
The best-proven use of a cannabis-based medicine is a purified CBD medicine (Epidyolex) for seizures in a few rare, severe childhood epilepsies. STRONGEvidence strength Purified CBD (Epidyolex), taken alongside another anti-seizure drug, reduces seizure frequency in Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex. It went through randomised controlled trials and is recommended by NICE for these specific conditions. Key source: NICE guideline NG144 recommends cannabidiol (with clobazam) for these epilepsies where seizures are inadequately controlled, subject to regular review of the response. Note the scope: this is a specific, purified, non-intoxicating medicine for specific rare epilepsies, not evidence that cannabis helps epilepsy in general.
Where it's a real but modest option: MS spasticity and chemo nausea
Two more uses have enough evidence to be offered on the NHS, but the benefit is more modest and second-line. They're tried when standard treatments haven't worked well enough.
Muscle spasticity in multiple sclerosis. A THC:CBD mouth spray (Sativex) can ease the stiffness and spasms of MS for some people. MIXEDEvidence strength NICE recommends offering a 4-week trial of the THC:CBD spray for moderate-to-severe MS spasticity when other treatments haven't helped, and continuing only if symptoms improve by at least 20%. So it helps some people meaningfully, but not everyone, and the effect is moderate rather than dramatic. Key source: NICE guideline NG144. Take it as: a worthwhile option to try under specialist care, with a clear "is it actually working?" check built in.
Nausea and vomiting from chemotherapy. A synthetic THC-like capsule (Nabilone) can help when standard anti-sickness drugs aren't enough. MIXEDEvidence strength NICE recommends considering nabilone as an add-on for chemotherapy-induced nausea and vomiting that persists despite conventional anti-sickness medicines. It's an add-on for a hard-to-treat problem, not a first choice. Key source: NICE guideline NG144.
Where the evidence is thinner than the reputation: chronic pain
Chronic pain is the single most common reason people seek medical cannabis privately, and it's also where the gap between hope and evidence is widest. THINEvidence strength NICE reviewed the evidence and recommends against offering cannabis-based products (nabilone, dronabinol, THC, or THC:CBD) for chronic pain, and says CBD should only be used for chronic pain within a clinical trial, because the benefit didn't justify the cost and risk. Key sources: NICE guideline NG144; and a 2025 living systematic review (AHRQ) which found only small improvements in pain and function from a balanced THC:CBD spray, at low-to-moderate certainty overall. Take it as: some people report real relief, and a specialist may still prescribe privately for an individual, but "cures chronic pain" is not what the evidence says. Expect a modest, uncertain effect, not a fix.
Where it's emerging or unproven: anxiety, sleep, PTSD and more
Anxiety, sleep problems and PTSD are among the most common reasons people want cannabis, but they're also where the medical evidence is weakest. THINEvidence strength There are no cannabis-based products licensed or recommended for anxiety, depression, insomnia or PTSD. The evidence is limited, mixed and mostly low-quality, and THC in particular can worsen anxiety in some people rather than help it. Context: much of the support here is patient-reported and observational rather than from robust randomised trials. Real experiences, but not the same as proof a product works for a condition in general. If this is you: the harm-reduction and quitting sides of this site (is it a problem? and using to cope) are worth a look, because self-treating anxiety or sleep with cannabis can quietly become its own issue.
A handful of other uses come up (appetite and wasting, some symptoms in palliative care, hard-to-treat conditions like Tourette's) where individual specialists may prescribe on a named-patient basis, but the formal evidence is similarly limited. The pattern across all of them is the same: promising stories, thin trial data.
The honest summary
Medical cannabis is not snake oil and it's not a miracle. It's a genuine, well-evidenced treatment for a few specific conditions, a reasonable second-line option for a couple more, and an uncertain bet for many of the things it's most hyped for. That's not a reason to dismiss it. It's a reason to go in with clear eyes, and to make any decision with a specialist doctor who knows your situation, not a website or a clinic's advertising. What that access actually looks like, in the UK and elsewhere, is on The Law Around the World.