Cannabis for Insomnia: Clinical Trials vs Product Claims

Walk into any dispensary and ask for “something to help me sleep,” and you’ll get a confident answer in seconds. A heavy indica. A CBN gummy marketed as “the sleepy cannabinoid.” A purple-tinged flower with a name like Nighttime or Knockout. The budtender means well, and the packaging looks like science.

But here’s the gap nobody at the counter mentions: the number of rigorous clinical trials testing cannabis specifically for insomnia disorder — the diagnosable medical condition — can be counted on two hands. And what those trials actually measured looks very different from what the marketing implies.

This article is not about the occasional restless night. We have other pieces for that, like Cannabis and Sleep: The Science Behind Better Rest and our science-based guide to the best strains for sleep. This one is narrower and more skeptical. It’s about clinical insomnia disorder. And it’s about the gap between what the randomized controlled trials (RCTs) found and what gets printed on a dispensary label.

A quick, important note before we go further. This is educational content, not medical advice. Insomnia disorder is a real clinical diagnosis. Self-medicating it — with cannabis or anything else — can mask an underlying condition. Please talk to your doctor or a sleep specialist before using cannabis for sleep. Nothing here is a cure, a treatment recommendation, or a substitute for professional care.

What “insomnia disorder” actually means

Most people use “insomnia” loosely to mean a bad night. Clinicians mean something much more specific. Insomnia disorder is diagnosed when someone has trouble falling asleep, staying asleep, or waking too early at least three nights a week for three months or longer. On top of that, the difficulty has to cause real daytime trouble — fatigue, mood problems, or trouble functioning.

That distinction matters a lot when you read research. A study showing cannabis helped people “sleep better” after a stressful week tells you almost nothing about chronic insomnia disorder. The chronic condition is stubborn. Research suggests it may overlap with anxiety, depression, pain, or circadian rhythm problems, and it tends to resist quick solutions. (Our piece on cannabis and sleep in depression digs into how messy that overlap gets.)

So when we evaluate the evidence, the question isn’t “does cannabis make you drowsy?” Of course it can. The question is: does it meaningfully treat the disorder, measured properly, over time, without trading one problem for another?

What the clinical trials actually show

Here’s the honest headline: there are only a handful of genuine RCTs in people with insomnia, and most are small, short, and report subjective improvement. Let’s walk through the ones that matter.

The ZTL-101 cannabinoid extract trial [Walsh et al., 2021]

This is the closest thing to a flagship insomnia RCT. Walsh and colleagues ran a randomized, double-blind, placebo-controlled crossover trial of a sublingual cannabinoid extract called ZTL-101 in adults with chronic insomnia symptoms [Walsh et al., 2021]. People took it nightly for two weeks, with a washout between arms.

The results look encouraging. Compared with placebo, the extract appeared to lower Insomnia Severity Index scores by about 5 points (a meaningful shift), to cut self-reported time to fall asleep, and to add total sleep time. It also seemed to move some objective actigraphy measures: wake-after-sleep-onset dropped and sleep efficiency rose. Side effects were mostly mild and cleared by morning.

That’s a real signal. But read the fine print: n = 23. Twenty-three people. Two weeks. It was a multi-cannabinoid extract — not flower, not a gummy, not “an indica.” The authors framed it as a promising proof-of-concept, not a green light to grab whatever’s labeled “sleepy” at the shop.

The CBN trial everyone cites but few have read [Bonn-Miller et al., 2024]

CBN (cannabinol) is the cannabinoid the sleep-product industry built a marketing empire on. So what does the largest RCT actually say? Bonn-Miller and colleagues ran a double-blind, placebo-controlled trial in 293 adults with poor self-reported sleep [Bonn-Miller et al., 2024]. They tested 20 mg CBN alone, plus CBN combined with various CBD doses, over seven nights.

The finding was nuanced. The 20 mg CBN dose was linked to fewer nighttime awakenings and less overall sleep disturbance versus placebo. But it did not seem to help people fall asleep faster, and it did not move wake-after-sleep-onset or daytime fatigue. Adding CBD didn’t help either. So CBN may do something modest. It’s a far cry from the “knocks you out cold” framing on the packaging. (We unpack the hype in CBN: The Sleepy Cannabinoid — Facts vs Fiction.)

Nabilone for PTSD nightmares [Jetly et al., 2015]

Nabilone is a synthetic THC analog. A small randomized, double-blind, placebo-controlled crossover study reported that it reduced the frequency and intensity of nightmares in service members with PTSD [Jetly et al., 2015]. That’s a specific, valuable result. But notice it’s about trauma-related nightmares, not everyday insomnia disorder, and the sample was tiny.

The systematic review that ties it together [Suraev et al., 2020]

When researchers pooled the whole field — preclinical and clinical — in a Sleep Medicine Reviews systematic review, the verdict was sobering [Suraev et al., 2020]: insufficient evidence to support routine clinical use of cannabinoids for any sleep disorder. Promising signals worth more trials, yes. A green light, no.

Subjective sleep vs objective sleep: the catch most labels ignore

Here’s the part that should make you cautious. Almost all the positive cannabis-sleep findings are subjective — people report sleeping better. That’s not nothing; how rested you feel matters. But subjective and objective sleep can diverge, and with cannabis they sometimes do in concerning ways.

THC tends to suppress REM sleep

THC is the intoxicating cannabinoid behind most of the “heavy indica” sedation. Research suggests it may suppress REM sleep — the dream-rich stage tied to memory and emotional processing [Kaul et al., 2021]. People often like this in the short term, because it can mean fewer vivid dreams and faster sleep onset. But REM isn’t filler. It’s functional sleep, and cutting it short night after night may carry costs we don’t fully understand yet. Our deep-dive on how THC affects REM sleep and dream recall covers this in detail.

Tolerance builds fast

The second catch is tolerance. Reviews of the research note that the sleep effects of THC tend to fade with repeated nightly use [Kaul et al., 2021]. The dose that knocked you out in week one may do much less by week six. That can push people toward larger doses — the opposite of what you want for a chronic condition. (This is the same tolerance machinery we explain in our complete guide to cannabis tolerance breaks; also relevant: why a T-break works.)

Rebound on cessation

Third: when chronic users stop, sleep often gets worse before it gets better — rebound insomnia, vivid “dream rebound” as suppressed REM comes flooding back, and restless nights. For someone using cannabis to manage insomnia disorder, that can create a dependency loop: the thing relieving the symptom becomes the thing the body now needs to avoid withdrawal-driven sleeplessness.

None of this means cannabis is useless for sleep. It means the simple story — “indica = sleep, problem solved” — leaves out the parts that matter most for a chronic condition managed over months.

Why dispensary claims outrun the evidence

So why the gap between thin RCT data and confident shelf labels? A few reasons worth naming.

“Indica = sedating” is folklore, not pharmacology. The indica/sativa split predicts almost nothing about effects. What you feel comes from the cannabinoid and terpene profile, not the plant’s leaf shape. We make this case at length in Indica vs Sativa vs Hybrid: What Science Actually Says in 2026 and the older indica vs sativa myth piece. A strain labeled “indica” might be myrcene-heavy and genuinely relaxing — or it might not be. The label alone is not evidence.

Terpene marketing runs ahead of the data. Myrcene gets called “the couch-lock terpene” — and turns up in products marketed for deep relaxation — and linalool (the lavender terpene) gets sold as a sleep aid. The mechanistic stories are plausible and there’s interesting science on pathways like adenosine, the brain’s sleep-pressure signal — see also how terpenes may relieve pain through adenosine receptors. But “plausible mechanism in a petri dish” is not “proven to treat insomnia disorder in humans.”

CBN is the clearest case of hype outrunning data. As we saw, the largest CBN trial found only modest effects on awakenings, nothing on falling asleep faster, and no boost from adding CBD. Yet “CBN = sleep cannabinoid” is printed on thousands of products. The claim got ahead of the evidence and never looked back. (It’s worth remembering CBD itself isn’t a straightforward sedative either — at low doses it can be stimulating rather than sleep-inducing.)

Subjective relief sells. Products are validated by reviews (“knocked me right out”), and those reviews capture the short-term, subjective, easily-tolerized effect — exactly the part the trials warn us not to over-read.

How to think about it (carefully)

If you and your doctor decide cannabis has a place in your sleep toolkit, a few evidence-aware principles:

• Treat it as short-term support, not a chronic solution. The data is strongest for brief use; tolerance and rebound are the long-game problems.
• Lowest effective dose. Higher THC isn’t better for sleep architecture and accelerates tolerance.
• Watch the REM tradeoff. If you notice flat mood, foggy memory, or you’ve stopped dreaming entirely, that’s worth discussing with your clinician.
• Track your own response. This is the entire philosophy behind High IQ: the “right” product for your insomnia is the one your own logged data says works, not the one the label promises. Patterns beat folklore.
• Don’t ignore the basics. Cannabis can’t out-muscle a wrecked sleep schedule, late caffeine, or untreated anxiety. CBT-I (cognitive behavioral therapy for insomnia) remains the first-line, evidence-backed treatment for the disorder — and cannabis is not a substitute for it.

Talk to your doctor — really

Insomnia disorder can be a symptom of something else: sleep apnea, depression, thyroid issues, or a medication side effect. Reaching for a sleep gummy can paper over a problem that needs a real diagnosis. So before you use cannabis to manage clinical insomnia, loop in a physician or sleep specialist. This matters most if you take other medications, are pregnant or nursing, or have a history of substance use or mental-health conditions. This article is information, not a prescription.

Key takeaways

• Only a handful of small, short RCTs have tested cannabis in people with insomnia, and most measured subjective improvement. The leading systematic review found insufficient evidence for routine clinical use.
• The strongest signal — the ZTL-101 extract trial — was promising but tiny (n = 23) and used a sublingual extract, not flower or “an indica.”
• CBN’s reputation outruns its data. The largest trial showed only modest effects on awakenings and nothing on falling asleep faster.
• Research suggests THC may suppress REM sleep, that tolerance to its sleep effects tends to build, and that stopping may trigger rebound insomnia.
• Treat cannabis as possible short-term support, not a chronic fix — and loop in a doctor first. CBT-I remains the first-line, evidence-backed treatment.

Frequently asked questions

Is there strong proof cannabis is an effective treatment for insomnia disorder? Not yet. A few small, mostly short RCTs show promising subjective results — and one, the ZTL-101 extract trial, showed some objective signal. But the leading systematic review concluded the evidence is insufficient for routine clinical use. Promising is not the same as proven.

Does CBN really make you sleepy? Modestly, at best. The largest RCT found 20 mg CBN reduced nighttime awakenings and overall sleep disturbance but did not help people fall asleep faster, and adding CBD didn’t help. The marketing is far louder than the data.

Is an “indica” the best choice for sleep? The indica/sativa label is a poor predictor of effects. Cannabinoid and terpene content matter more — and even those are still being studied for sleep specifically.

Will using cannabis nightly stop working? Often, yes. Tolerance to the sleep-promoting effects of THC tends to build with repeated use, and stopping can trigger rebound insomnia and vivid dream rebound. That’s a key reason it’s better suited to short-term support than chronic management.

Does it hurt my sleep quality even if I feel rested? Possibly. THC tends to suppress REM sleep, so you may feel you slept hard while losing dream-stage sleep that does important work. Feeling rested and having healthy sleep architecture aren’t the same thing.

Sources

• Walsh JH, Maddison KJ, Rankin T, et al. (2021). Treating insomnia symptoms with medicinal cannabis: a randomized, crossover trial of the efficacy of a cannabinoid medicine compared with placebo. Sleep, 44(11). DOI: 10.1093/sleep/zsab149
• Bonn-Miller MO, Feldner MT, Bynion TM, et al. (2024). A double-blind, randomized, placebo-controlled study of the safety and effects of CBN with and without CBD on sleep quality. Experimental and Clinical Psychopharmacology. DOI: 10.1037/pha0000682
• Jetly R, Heber A, Fraser G, Boisvert D. (2015). The efficacy of nabilone in the treatment of PTSD-associated nightmares: a preliminary randomized, double-blind, placebo-controlled cross-over design study. Psychoneuroendocrinology, 51. DOI: 10.1016/j.psyneuen.2014.11.002
• Suraev AS, Marshall NS, Vandrey R, et al. (2020). Cannabinoid therapies in the management of sleep disorders: a systematic review of preclinical and clinical studies. Sleep Medicine Reviews, 53. DOI: 10.1016/j.smrv.2020.101339
• Kaul M, Zee PC, Sahni AS. (2021). Effects of cannabinoids on sleep and their therapeutic potential for sleep disorders. Neurotherapeutics, 18(1). DOI: 10.1007/s13311-021-01013-w