CBN for sleep: a new cannabinoid or just hype? What the studies say 2026

CBN, or cannabinol, has been marketed for several years as the "sleep cannabinoid." As of 2021, there was not a single clinical study that tested this description with sleep recording or validated surveys. By 2026, four randomized trials involving humans had been published, so for the first time, data can answer the question rather than relying on analogies to THC or results from rodents. The answer is more modest than the labels suggest: in none of these trials did CBN perform better than a standard melatonin tablet, and the study with the most rigorous measurement did not achieve its primary endpoint. Below, we break down these four trials: who conducted them, at what doses, and what exactly was measured.

KEY INFORMATION
• Corroon's review (Cannabis and Cannabinoid Research, 2021) sifted through 99 studies on humans and found no trial measuring sleep after CBN with polysomnographic recording.
• The first such trial (Lavender et al., Journal of Sleep Research, 2026) did not achieve its primary endpoint.
• In two large trials, CBN did not perform better than 4-5 mg of melatonin.
• The opinion about sedation comes from a 1975 study on five people, in which CBN alone did not induce sleep.
• CBN is formed from the oxidation of THC, so its presence mainly indicates the age of the raw material.

Where does CBN come from in sleep products?

CBN is not a compound that cannabis produces in larger quantities. It is formed from THC under the influence of oxygen, heat, and light, so its concentration increases with the age of the raw material. In hemp resin stored for two years, THC content dropped from 35.2% to 2.7%, while CBN increased from 0.9% to 6.9% (Fettoukh i in., Scientific Reports, 2025).

This one dependency explains most of the misunderstandings surrounding cannabinol. A high CBN content in a sample is not a characteristic of the strain or a credit to the grower. It is a record of storage history: how many months the material was stored and under what conditions. The authors of that study used this directly, building a model that estimates the aging time of a sample based solely on the cannabinoid profile with sufficient accuracy for legal purposes.

A producer who wants to put a specific number of milligrams of CBN on the label has three options. They can start with old plant material, deliberately oxidize THC under controlled conditions, or purchase an isolate obtained through chemical conversion. None of these options align with what the word "plant-based" on the packaging suggests, and none are problematic in themselves. The problem arises when conclusions about its sleep-inducing effects are drawn from the origin of the molecule. That is a separate question and has a separate answer.

Where did the opinion that CBN induces sleep come from?

From a study conducted in 1975 with five volunteers. Karniol and colleagues administered orally, at weekly intervals, a placebo, 50 mg of CBN, 25 mg of THC, and both substances together. After CBN alone, participants did not report drowsiness or intoxication. They reported these effects after THC, and most strongly after the combination of THC with CBN (Karniol i in., Pharmacology, 1975).

This is the opposite of what marketing claims. In that experiment, cannabinol did not induce sleep on its own but enhanced the effect of THC, and to a degree that the authors themselves assessed as minor. Their conclusion was that such an interaction does not explain the greater potency of the plant material. No one studied sleep at that time. Heart rate, pain threshold, and time perception were measured, and feelings were collected from a list of 66 adjectives.

The second hypothesis suggests that the sedation attributed to old extracts was due to the terpenes that remained in them. Russo (British Journal of Pharmacology, 2011) It gathered arguments that myrcene and linalool have their own calming effects, independent of cannabinoids. This is still a hypothesis, not a result from an experiment separating both components in humans. We write more about these compounds in the text Terpeny w CBD.

The scale of the phenomenon is real. In a representative survey among adult Americans, 4.5% have ever tried CBN, with insomnia being the most commonly cited medical reason, at 15.4% (Satybaldiyeva i in., Journal of Cannabis Research, 2025).

How many studies on CBN in humans will there be in 2026?

Four randomized placebo-controlled trials, all from 2024-2026. There were none before that. Corroon's review sifted through the abstracts of 99 studies on humans, read eight full papers that met the criteria, and found not a single trial assessing sleep using polysomnography or a validated sleep questionnaire (Corroon, Cannabis and Cannabinoid Research, 2021).

Four trials are still very few for a substance that has been sold for years as a sleep aid. The following summary shows exactly what was measured in them and what the results were.

Three out of four trials based their sleep measurement solely on a questionnaire filled out by participants, and two were built around a specific commercial product from one manufacturer. This does not invalidate the results but calls for caution, as the questionnaire also captures the expectations of the subject, and placebo effects in sleep studies are usually strong.

What did the CBN study with polysomnography show?

That CBN did not shorten the wake time after falling asleep. In the CUPID trial, twenty individuals diagnosed with insomnia received 30 mg of CBN, 300 mg of CBN, and placebo in random order, each time for one night under full sleep recording. The primary endpoint was not met at any dose (Lavender i in., Journal of Sleep Research, 2026).

The additional results look more interesting. At 300 mg, sleep in NREM-2 stage was prolonged, the time to fall asleep was shortened, the number of awakenings visible in the EEG recording decreased, and the subjective quality of sleep improved. Participants reported a total of 247 adverse events of mild to moderate severity, counting all arms of the study together.

This is what an honest reading of the data looks like. Additional points in a study that did not reach the primary endpoint pose a hypothesis rather than confirm it. The authors state this directly: larger and longer trials are needed. Also note at what dose anything changed. That is 300 mg at once, which is a multiple of what you will find in a gummy or a drop of oil marketed with the sleep cannabinoid slogan. At 30 mg, none of these effects were recorded.

Does CBN perform better than melatonin?

There is no study that shows this. In the trial by Kolobaric and colleagues, 1020 people received CBN for four weeks at doses of 25, 50, or 100 mg, 4 mg of melatonin, or placebo. All active groups improved compared to placebo, but no dose of CBN performed better than melatonin (Kolobaric i in., Pharmaceuticals, 2024).

A second study from the same group included 1793 people, also for four weeks. Adding 15 mg of CBN to 15 mg of CBD did not improve the outcome compared to CBD alone, and none of the combinations studied surpassed 5 mg of melatonin (Saleska i in., Journal of the American Nutrition Association, 2024).

The third trial yielded mixed results. 293 people took 20 mg of CBN for seven nights, either alone or with CBD. The primary endpoint, which was overall sleep quality, did not achieve statistical significance. However, two additional points emerged: fewer awakenings at night and a lower rate of sleep disturbances. Time to fall asleep, wakefulness after falling asleep, and daytime drowsiness did not change at all, and adding CBD did not improve anything (Bonn-Miller i in., Experimental and Clinical Psychopharmacology, 2024).

The common denominator of these trials is this: CBN is somewhere around melatonin, never above it. With the price of a dose being multiple times higher, this is information that should suffice for a purchasing decision.

Is it worth combining CBN with melatonin in one product?

Such products are on the market, but they have a pharmacokinetic problem that the labels do not mention. CBN inhibits the CYP1A2 enzyme, which is the main liver enzyme that breaks down melatonin. In mice, the administration of cannabinol increased blood exposure to melatonin fourfold (Anderson i in., Basic and Clinical Pharmacology and Toxicology, 2025).

This is a preclinical study, and this number should not be directly transferred to humans. But the direction of the phenomenon arises from the mechanism, not from guessing. The same group previously described the inhibition of caffeine metabolism by CBN, which breaks down via the same pathway. The practical conclusion is simple: a product combining CBN with melatonin may act like a higher dose of melatonin, rather than as two independent components.

For someone who was already looking for melatonin, this is not a disaster, but a loss of control over the dosage. However, with melatonin, it's not just about how much, but also at what time, and both of these variables are easier to manage when you take it separately. We describe how to do this in the text. Melatonina na sen. If you are taking medications metabolized by CYP1A2, discuss adding CBN to your evening regimen with your doctor or pharmacist beforehand.

Does the CBN dosage on the label correspond to the dosages from studies?

Rarely. Corroon pointed out that products sold for sleep usually contain up to 5 mg of CBN per serving, and he directly recommended that future studies use significantly higher doses. The trials that have emerged since then have done just that: 15, 20, 25, 50, 100, and 300 mg.

The difference can be sixtyfold. The signals that did appear came at 20 mg and above, with the clearest signals at 300 mg. A label with 5 mg does not replicate any of these conditions, even if the stated number is accurate.

A separate question is whether it is accurate. No study has been published for CBN checking the compliance of declarations with actual content, as was done for CBD oils and flower. The only thing you can verify yourself is the certificate of analysis for a given batch from an external laboratory, stating the CBN content in milligrams per serving. Without this document, the number on the package remains a declaration from the manufacturer, not a measurement.

What has stronger foundations than CBN if you want to sleep better?

Melatonin in a small dose and a structured evening. In the only trial that had a placebo arm, 4 mg of melatonin improved sleep compared to placebo, and no dose of CBN did better. Melatonin is also cheaper per serving and has a well-documented usage pattern.

Before you start looking for another molecule, check the conditions under which it would work. A consistent wake-up time, even on weekends. A bedroom cooler than the rest of the apartment. No caffeine in the afternoon and an evening without a bright screen. It's boring, free, and addresses the cause rather than the symptom. Our experience with reader questions is that those reaching for exotic cannabinoids usually haven't gone through this stage yet.

If you are interested in the cannabinoid part of the puzzle, a better-documented starting point is CBD, which we discuss in the text. Czy CBD pomaga na sen. It has more studies and a clearer safety profile than cannabinol.

So the answer to the titular question is: CBN is not a scam; it is a poorly studied substance sold in doses lower than those that have been tested at all, and without proven advantages over a cheaper alternative. That's not enough to start with it.

Frequently Asked Questions

Czy CBN pomaga na sen?

The data is weak. By 2026, four randomized trials of CBN in humans had been published. The only one with sleep recording (Lavender i in., Journal of Sleep Research, 2026, 20 people) did not achieve the primary endpoint, meaning it did not shorten the time awake after falling asleep. The other three relied on surveys, and none showed an advantage of CBN over melatonin.

Does CBN work better than melatonin?

Not in any published study. In the trial Kolobaric i in. (Pharmaceuticals, 2024) 1,020 people took CBN at doses of 25, 50, or 100 mg, 4 mg of melatonin, or placebo for four weeks. All active groups improved compared to placebo, but no difference between CBN and melatonin was found at any dose.

What dose of CBN was studied in clinical trials?

From 15 to 300 mg. Saleska et al. used 15 mg, Bonn-Miller et al. 20 mg, Kolobaric et al. 25, 50, and 100 mg, and Lavender et al. 30 and 300 mg. Corroon (2021) pointed out that commercial sleep products usually contain up to 5 mg of CBN per serving, which is less than any of the studied doses.

Czy sam CBN usypia, bez THC?

In the study that started this opinion, it did not induce sleep. Karniol i in. (Pharmacology, 1975) they administered 50 mg of CBN, 25 mg of THC, and both substances together to five volunteers. Drowsiness and intoxication were reported after THC and after the combination of both, not after CBN alone. The effect attributed to cannabinol was a result of the mixture.

Czy CBN jest bezpieczny?

Short-term, it appears to be well tolerated, but data is limited. In four-week trials with doses up to 100 mg, the incidence of adverse effects did not differ from placebo. In a trial with 300 mg in one night, 247 events of mild to moderate severity were recorded across all arms. There is a lack of data beyond a few weeks of use.

Where does CBN come from in products?

From the oxidation of THC. Cannabis does not produce CBN in larger quantities; it is formed from THC under the influence of oxygen, heat, and light. In resin stored for two years, THC dropped from 35.2% to 2.7%, while CBN increased from 0.9% to 6.9% (Fettoukh i in., Scientific Reports, 2025). Producers resort to old material, intentional oxidation, or chemical conversion.

At u Bucha, there are no products with CBN, so we have nothing to sell you under this slogan. Evening support with stronger foundations can be found in the section supplements.

This article is for informational and educational purposes and does not constitute medical advice. Before starting to use cannabis or CBD for therapeutic purposes, consult a doctor, especially if you are taking other medications, are pregnant, or breastfeeding.

Author: Michał Waluk · Opublikowano: 2026-06-22 · Aktualizacja: 2026-08-08