Daria, 44, has generalized anxiety disorder and has used cannabis nightly for two years to help her fall asleep. She tells you it’s the only thing that works. “And it’s all natural,” she adds, “so I’m not worried about it.” Upon further questioning, she acknowledges needing more than before and being unable to sleep when she skips it.

Cannabis is one of the most common self-treatments for insomnia (Morin CM and Buysse DJ, New Engl J Med 2024;391(3):247–258). As legalization expands, so does the belief that cannabis is a safe, natural alternative to prescription sleep aids. You’ve probably heard patients tell you that it’s “better than Ambien.” Whether that’s true is not so straightforward, and helping patients understand why can lead to a useful clinical conversation.

Here, we’ll review the relationship between cannabis and sleep, including acute effects, chronic consequences, and withdrawal-related insomnia, as well as how to discuss sleep and cannabis with your patients.

Short-term use: Some (subjective) benefits
THC, the primary psychoactive component of cannabis, may have some sleep benefits in the short term. Some studies show that THC may reduce sleep latency, or the time it takes to fall asleep (Gorelick DA, New Engl J Med 2023;389(24):2267–2275). It also increases subjective sleepiness and may increase slow-wave (deep) sleep early in use (Gonzalez JE et al, J Sleep Res 2026;35(4):e70298).

While these effects may improve sleep subjectively, studies suggest that THC, particularly at higher doses, can reduce rapid eye movement (REM) sleep (Suraev A et al, J Sleep Res 2026;35(1)e:70124). REM sleep is linked to dreaming, emotional memory processing, and mood regulation. So, subjective and objective sleep quality may diverge; patients may feel like they’re sleeping well while their sleep architecture is significantly compromised.

Chronic use: Tolerance and escalation
The sedative effects of cannabis are not durable. Tolerance to THC’s sleep-promoting properties develops quickly, often within weeks; as tolerance builds, patients use more to achieve the same effect, often without recognizing the gradual escalation (Gorelick, 2023). Chronic use may also degrade sleep quality even further. Over time, heavy regular use is associated with:

• Fragmented sleep
• Reduced total sleep time
• Increased nighttime awakenings

(Source: Mao F et al, Sleep Med Rev 2025;84:102189)

Regular use can also produce physiologic dependence, setting the stage for withdrawal-related insomnia when patients try to stop (Connor JP et al, Addiction 2022;117(7):2075–2095).

Withdrawal insomnia
Cannabis withdrawal syndrome is recognized in DSM-5 and includes:

• Irritability
• Anxiety
• Decreased appetite
• Restlessness
• Insomnia

Withdrawal symptoms typically begin within 24–48 hours of stopping, peak at 2–6 days, and can persist for up to 3 weeks (Budney AJ et al, J Abnorm Psychol 2003;112(3):393–402). For many patients, sleep disruption may be the hardest withdrawal symptom to tolerate.

A notable feature of cannabis withdrawal is REM rebound. After weeks or months of REM suppression, stopping cannabis can trigger a surge in REM sleep, which often manifests as vivid or disturbing dreams (Velzeboer R et al, Sleep Med Rev 2025;84:102164). If patients plan to decrease or stop using cannabis, forewarn them about this phenomenon and reassure them that it is temporary. For more on cannabis withdrawal, see the interview with Dr. Alan Budney in CATR Jan/Feb/Mar 2025.

What about CBD?
CBD, the primary non-intoxicating component of cannabis, has attracted substantial interest as a sleep aid. But the evidence is far thinner than the marketing suggests. Most studies are small, use heterogeneous formulations, and lack placebo controls. Effects appear dose-dependent and inconsistent, and many commercial CBD products contain little of what’s on the label (www.tinyurl.com/epcudrhb).

Unlike THC, CBD does not appear to suppress REM sleep, and it may have anxiolytic effects at higher doses that indirectly support sleep in patients with anxiety disorders (Kansagara D et al, JAMA Intern Med 2026;186(5):618–628). However, there is currently no robust evidence that CBD improves sleep in people without an underlying anxiety disorder. For most patients, when it comes to CBD, the honest answer is that the evidence isn’t there yet, plus the products are unregulated and may not contain what they claim (Gardener H et al, Sci Total Environ 2022;851(Pt 1):158110).

Special populations worth flagging

Patients with PTSD
These patients often use cannabis specifically to suppress nightmares, and THC’s REM suppression does blunt nighttime trauma re-experiencing in the short term (LaFrance EM et al, J Affect Disord 2020;274:298–304). However, regular use can delay psychological processing of trauma and worsen trauma-related symptoms in the long term (Metrik J et al, Psychol Med 2022;52(3):446–456). Established PTSD treatments, not REM suppression, are the durable solution.

Adolescents and young adults
Adolescents and young adults may be particularly vulnerable to cannabis-related sleep disruptions with regular use (Coelho J et al, Psychiatry Res 2023;322:115096). The developing brain is particularly vulnerable to long-term adverse effects of cannabis exposure, including increased risk of psychotic, bipolar, depressive, and anxiety disorders, with stronger associations at younger ages of use (Young-Wolff KC et al, JAMA Health Forum 2026;7(2):e256839).

Older adults
Older adults are also more susceptible to negative effects from cannabis, albeit for different reasons than young people. Age-related physiological changes (reduced hepatic function, polypharmacy, CNS sensitivity, reduced lean body mass) increase the risk of adverse effects, particularly sedation and falls (Hsu M et al, JAMA 2026;335(4):345–359).

Patients with psychiatric comorbidities
Patients with psychiatric disorders frequently use cannabis for sleep-related issues (Kuhathasan N et al, BMC Psychiatry 2022;22(1):303). However, cannabis is associated with worse outcomes across a range of psychiatric conditions, including anxiety, mood, and particularly psychotic disorders (Kansagara et al, 2026).

Pregnancy and lactation
Counsel pregnant patients to discontinue cannabis entirely as it is associated with low birth weight, preterm birth, perinatal mortality, and need for NICU admission. THC is passed into breastmilk as well, so though use of cannabis is not entirely contraindicated while breastfeeding, we should still recommend that patients stop using it (Valent AM et al, Obstetrics and Gynecol 2025;146(4):600–611).

How to have the conversation
Most patients using cannabis for sleep believe it is helping them and have not been told otherwise. Approach the conversation as information-sharing, not confrontation.

First, validate the patient’s experience: “You’re right that cannabis can help you fall asleep faster and that, in the short term, it can feel like your sleep is improved. The problem is what happens to sleep quality over time.” Explain tolerance and REM suppression: “With regular use, the brain adjusts, and you need more to get the same effect. Meanwhile, your sleep quality is gradually worsening, even if it doesn’t feel that way day to day.”

When assessing cannabis-related insomnia, distinguish between primary insomnia and cannabis-induced sleep disturbance by asking these simple questions:

• Are symptoms worse without cannabis?
• Has the dose increased over time?

A “yes” to either question suggests cannabis is contributing to poor sleep, not just a response to it.

For patients willing to cut back or stop, set expectations from the outset. Sleep will likely worsen before improving, and the first few weeks will be the hardest. Vivid, even disturbing, dreams should be expected.

While the question has not been rigorously studied, a short-term sleep medication is reasonable to help patients get through the early stages of withdrawal (see article on page 1 for more about non-addictive hypnotic medications).

Treating insomnia sustainably
Start with foundational measures: Avoid caffeine before bed, practice good sleep hygiene, and treat comorbid psychiatric conditions.

Hypnotic medications can be useful short term but carry risks. Benzodiazepines and Z-drugs can be habit forming and are associated with dependence, withdrawal, falls, cognitive impairment, and daytime sedation. Like cannabis, these medications may improve subjective sleep while disrupting normal sleep architecture (Barbaux L et al, Sleep 2025; 48(10):zsaf168).

Utilize cognitive behavioral therapy for insomnia (CBT-i)
The first-line and best-supported treatment for insomnia is CBT-i, which has stronger long-term outcomes than any pharmacologic option, including cannabis (Trauer JM et al, Ann Intern Med 2015;163(3):191–204). Access remains the biggest barrier. Consider programs like Sleepio (sleepio.bighealth.com) or the VA’s free CBT-i Coach (www.tinyurl.com/582cmp8t) when therapist-delivered CBT-i is unavailable (Morin and Buysse, 2024).

When patients ask whether they’ll ever sleep without cannabis, be honest: It takes time. Most people see meaningful improvement within a few weeks, though full normalization of sleep architecture may take months. Framing this challenge as a known recovery trajectory can help patients stay the course.

For Daria, the clinical picture is consistent with cannabis-induced sleep disturbance superimposed on anxiety-related insomnia. The next step is not simply to tell her to stop, but to help her understand what cannabis is doing to her sleep and build a treatment plan that addresses her insomnia sustainably.

CARLAT VERDICT
Cannabis shortens sleep onset but disrupts sleep quality over time by suppressing REM sleep and producing withdrawal insomnia that drives continued use. Counsel patients nonjudgmentally, prepare them for a few tough weeks if they decide to stop, and build a comprehensive sleep plan consisting of psychoeducation, treatment of comorbid conditions, potentially a short-term hypnotic medication, and (if possible) CBT-i.