Sleep claims and evidence

Short-term findings: modest sleep improvements

Clinical trials show that medicinal cannabis can produce small, short-term improvements in sleep quality for some people. A 2021 systematic review in BMJ Open found that cannabis-based medicines reduced sleep disturbance scores by an average of 2-3 points on standardised scales over 2-4 weeks, compared with placebo. However, the review noted that most studies were small and short, and the effect size was not large enough to recommend cannabis as a first-line sleep treatment.

The strongest evidence comes from studies of nabiximols (a 1:1 THC:CBD spray) and nabilone (a synthetic THC analogue), both of which showed reduced time to fall asleep and fewer night-time awakenings in people with chronic pain or multiple sclerosis. For example, a 2014 trial in Journal of Pain reported that nabilone improved sleep quality in fibromyalgia participants over 8 weeks, but the benefit diminished after discontinuation.

Indirect pain effects: sleep as a secondary gain

Much of the sleep benefit attributed to medicinal cannabis is indirect, arising from pain relief rather than direct sedation. People with chronic pain often report that better pain control at night leads to fewer sleep disruptions. A 2020 analysis in Pain Medicine found that 60% of individuals who used cannabis for pain also reported improved sleep, but the effect was strongly correlated with pain reduction, not with THC dose.

This indirect pathway means that sleep improvement is not guaranteed. Those whose pain is not well-controlled by cannabis may see no sleep benefit, and people using cannabis specifically for insomnia without pain may have weaker results. A 2019 review in Current Psychiatry Reports concluded that the evidence for cannabis as a direct hypnotic is insufficient, and that any sleep effect is likely mediated by pain, anxiety, or muscle spasm relief.

Tolerance and diminishing returns

With regular use, the sleep-promoting effects of THC tend to diminish due to tolerance. A 2018 study in Sleep found that daily cannabis users showed less slow-wave sleep (deep sleep) after 3 weeks of continuous use, and their subjective sleep quality returned to baseline after 1-2 weeks. This suggests that cannabis may help sleep acutely but loses efficacy with daily use, potentially leading people to increase their dose to maintain the effect, a pattern that can escalate tolerance and dependence.

The NZ Formulary notes that tolerance to the sedative effects of THC develops within days to weeks, and advises prescribers to consider cycling or intermittent use to preserve efficacy. Anyone using cannabis for sleep should be aware that nightly use may lead to reduced benefit over time, and that “drug holidays” (periods of non-use) may be needed to reset sensitivity.

Next-day impairment: a significant risk

THC has a long half-life, 1-3 days in occasional users, longer in regular users, and its psychoactive effects can persist into the next day. A 2020 meta-analysis in Neuroscience & Biobehavioral Reviews found that cannabis use impairs next-day cognitive performance, including attention, memory, and reaction time, even when the user no longer feels “high.” This is especially relevant for people who take cannabis at night: they may wake up feeling groggy, with reduced ability to drive, operate machinery, or perform complex tasks.

The NZ Transport Agency advises that driving with any detectable THC in blood is illegal, and a prescription is not a defence. People should be counselled not to drive for at least 8-12 hours after using THC, and longer for high doses or oils, which have a delayed peak. Next-day impairment is dose-dependent: higher THC doses and oil-based products (which produce active metabolites that last longer) increase the risk.

Evidence limits: what we still don’t know

The evidence base for cannabis and sleep has significant gaps. Most studies are short (2-8 weeks), so long-term safety and efficacy are unknown. Many trials use synthetic cannabinoids (nabilone, dronabinol) rather than whole-plant products, making it hard to generalise to prescribed flower or oils. Few studies have examined sleep architecture (the stages of sleep) using polysomnography, so the effect on REM sleep and deep sleep is poorly characterised.

A 2022 review in Cannabis and Cannabinoid Research noted that CBD may have alerting properties at low doses and sedating properties at high doses, but the dose-response curve is not well defined. This means that a product labelled for sleep may actually disrupt sleep in some people, depending on the CBD:THC ratio and individual metabolism.

How sleep claims and evidence are assessed

Evaluating sleep claims requires looking at the type of study, the population studied, the outcome measured and the duration of follow-up. The method used across the research cited here draws on several checkable criteria. First, whether the trial used a placebo control and blinded participants and assessors. Second, whether sleep was measured by a validated scale (such as the Pittsburgh Sleep Quality Index) or by polysomnography, which directly records brain waves, eye movements and muscle activity during sleep. Third, whether the effect was statistically significant and clinically meaningful, meaning the size of the improvement mattered to daily function. Fourth, whether the benefit persisted beyond 4 weeks. Most cannabis sleep studies meet the first two criteria but not the last two, which is why the overall evidence is described as modest and short-term. A further consideration is whether the study examined people with a primary sleep disorder (such as insomnia) or people whose sleep problems were secondary to another condition (such as chronic pain). This distinction matters because the mechanism of any sleep benefit differs, and the evidence is weaker for primary insomnia.

Medicinal cannabis in New Zealand: regulatory context

Medicinal cannabis in New Zealand is available on prescription under the Misuse of Drugs (Medicinal Cannabis) Regulations 2020. Products must meet minimum quality standards set by the Medicinal Cannabis Agency. Prescribers must assess each person individually, and cannabis is not approved as a first-line treatment for any condition, including sleep disorders. The NZ Formulary states that cannabis-based medicines should be considered only when standard treatments have been tried and found ineffective or poorly tolerated. For sleep, this means options such as cognitive behavioural therapy for insomnia (CBT-I), melatonin, or addressing underlying causes should be exhausted first. Prescribers are also required to obtain informed consent, document treatment goals, and review progress at regular intervals. The Medicinal Cannabis Agency does not endorse specific products for sleep, and any claim that a particular cannabis product is “approved for sleep” in New Zealand is inaccurate. Products are prescribed off-label, and the responsibility for assessing the balance of benefit and harm rests with the prescribing doctor.

A practical guide to sleep claims and evidence

When you encounter a claim about cannabis and sleep, whether from a product website, a clinic advertisement or a news report, apply a short decision rule. Ask three questions. One: does the claim cite a specific study, and if so, what was the study design? A randomised controlled trial carries more weight than a survey or anecdote. Two: what was the primary outcome? If sleep was a secondary endpoint in a pain or anxiety trial, the claim should reflect that indirect relationship. Three: what is the time frame? A benefit shown at 2 weeks may not hold at 8 weeks, and tolerance is well documented. If a claim does not answer these three questions clearly, treat it with caution. The strongest claims are those backed by placebo-controlled trials that measured sleep as a primary outcome, used a validated scale or polysomnography, and followed participants for at least 4 weeks. No cannabis product currently marketed in New Zealand meets all three of those standards for a primary sleep indication.

Practical implications

Given the current evidence, anyone considering medicinal cannabis for sleep should:

• Discuss with their prescriber whether sleep is the primary target or a secondary benefit of pain or anxiety treatment.
• Start with a low THC dose (2.5-5 mg) and titrate slowly, noting that higher doses may worsen next-day impairment without improving sleep.
• Consider intermittent use (e.g., 3-4 nights per week) to reduce tolerance.
• Avoid driving or operating machinery for at least 12 hours after use, especially with oils.
• Monitor for tolerance: if the same dose stops working after 1-2 weeks, a drug holiday or dose adjustment may be needed.

The NZ Formulary recommends that cannabis for sleep be used only when first-line treatments (e.g., cognitive behavioural therapy, melatonin) have failed, and that people be reassessed after 4 weeks to determine if the benefit outweighs the risks.