Are Sweet Dreams Made of These? Understanding the Relationship Between Sleep and Cannabis Use

Abstract

Introduction: There is widespread literature on the interaction between cannabis use and sleep, yet the mechanisms that underlie this relationship are not well understood. Several factors lead to inconsistencies in this relationship suggesting a nuanced interaction between cannabis and sleep. An important question that remains to be addressed is the temporal relationship between disrupted sleep and cannabis use. This literature review summarizes the existing literature on the association between disrupted sleep and cannabis toward the goal of addressing the question of the chronology of these reported effects.

Materials and Methods: We conducted a review of the literature using PubMed to summarize current knowledge on the association between cannabis use and sleep in humans.

Results: We identified 31 studies on the association of cannabis use and sleep. The findings from these studies were mixed. Cannabis was associated with a variety of impacts on sleep ranging from beneficial effects, such as reduced sleep-onset latency, to negative outcomes, such as reduced sleep duration and suppressed rapid eye movement oscillations. The chronology of the interaction of cannabis and sleep was unclear, although much of the current literature focus on factors that modulate how cannabis impairs sleep after initial use.

Conclusion: There was sufficient evidence to suggest that cannabis use alters circadian rhythms, and hence, negatively impacts sleep. The current literature is largely from studies utilizing self-report measures of sleep; thus, objective measures of sleep are needed. In addition, although there were no empirical studies on the temporal relationship between cannabis use and sleep, the majority of the literature focused on characterizing sleep impairment after cannabis use.

Introduction

Cannabis is the most commonly used illicit drug with 3.8% of adults reporting using cannabis worldwide.¹ Consequently, the prevalence of cannabis use disorder has also risen from 1.5% to 2.9% from 2001 to 2013.² In the United States, cannabis has been legalized for recreational use in 15 states and medicinal use in 35 states. Recreational use is often defined as use of illicit products containing often high and uncontrolled levels of tetrahydrocannabinol (THC), whereas medicinal use refers to use of regulated products containing controlled levels of THC and/or cannabidiol (CBD).

Cannabis acts on the brain endocannabinoid system (ECS) that is involved in circadian rhythm modulation and sleep.³ Administration of cannabinoids promotes sleep⁴ through the activation of cannabinoid 1 (CB1) receptors,⁵ although various cannabinoids (e.g., THC and CBD) have varying affinities for CB1 receptors.^6,7 In an animal study, injection of CB1 antagonists inhibited sleep recovery after sleep deprivation suggesting that CB1 receptors within the ECS play a role in homeostatic sleep processes.⁸ It is hypothesized that the action of endocannabinoids on CB1 receptors triggers the release of sleep promoting neurotransmitters such as adenosine.⁹ Given that cannabis acts on the brain ECS that is involved in regulatory processes such as sleep,³ it is critical to understand the relationship between cannabis and sleep.

Sleep is an essential function that plays a key role in the regulation of many neurobiological functions such as neuronal communication.¹⁰ Research recommends 7–9 h per day for a healthy sleep schedule with any deviations resulting in higher risk for disease and disorder.¹¹ Sleep disruption or sleep impairment is generally defined as any alteration in the healthy homeostatic processes that is most commonly characterized by longer sleep-onset latency, changes in rapid eye movement (REM)/slow-wave sleep oscillations, increased limb movements, alterations in sleep duration or habitual sleep habits, or an increase in waking events during sleep.¹² Disruptions in sleep that include changes in regular sleep patterns that alters sleep–wake schedule, sleep stages, sleep duration, and sleep patterns have been associated with negative outcomes from objective and subjective measure of cognitive functioning.¹³ For example, 72 h of sleep deprivation in healthy adults has been associated with increased levels of cortisol and a significant reduction of cell proliferation in the hippocampus.¹⁴ In addition, sleep deprivation and variation in the sleep–wake cycle have been linked to altered synaptic plasticity, reductions in cerebral metabolism, and gray matter loss in brain structures like the hippocampus.^14,15

Given the role of sleep in facilitating neuronal communication, it is not surprising that sleep disruption has negative impacts on brain structure and function. Although the exact mechanisms have yet to be determined, chronic sleep disruption has been associated with reductions in gray matter volume in prefrontal cortical areas and the hippocampus^15–17—areas important for memory and executive functioning. In accord, sleep disturbances have also been shown to impair memory and executive functioning.^18,19 Finally, sleep also impacts mood such that shorter sleep duration has been associated with greater symptoms of depression in both children and adults.^16,20

The sleep literature on cannabis has relied predominantly on subjective measures given the relative ease of collecting these types of data compared with objective measures (Table 1A). A meta-analysis by Irwin et al.²¹ found that the most commonly used methods to evaluate sleep disturbances included single survey items, multiple symptoms reporting, validated questionnaires, or diagnosis.²¹ One of the most widely utilized questionnaires is the Pittsburgh Sleep Quality Index (PSQI).²² The PSQI consists of 19 items that assess sleep quality and sleep disturbances over the past month. Typically, global PSQI scores >5 are considered to indicate relevant sleep disturbances.²² Global PSQI scores for patients with insomnia have been shown to have a high correlation coefficient for test–retest reliability of 0.87.²³ Although sleep outcomes measured by these self-report methods are germane facets of sleep quality, the ability of these measures to fully capture underlying mechanisms may be limited.

Table 1.

Studies of Cannabis and Sleep in Human Participants Using (A) Subjective and (B) Objective Measures in Ascending Chronological Order of Publication

Year	Authors	Participant characteristics	Sleep assessment	Results
A. Studies using self-report measures
2004	Haney et al.46	7 male adults cannabis users Mean age=24 Cannabis administration—10 mg oral THC	St Mary's Hospital Sleep Questionnaire (modified version)	Active THC administration decreased sleep-onset latency. Abstinence decreased sleep duration
2008	Budney et al.49	67 adult cannabis users Mean age=31.9 Substantial daily cannabis user for at least 6 months before quit attempt	WDS	>50% of the cannabis group reported sleep that contributed toward a failed attempt to quit
2008	Roane and Taylor69	4,494 adolescents Mean age (baseline)=15.83	In-house insomnia questionnaire	Adolescents/young adults with insomnia symptoms were significantly more likely to use cannabis. Men were more likely to endorse cannabis use
2008	Vandrey et al.45	12 adult cannabis and tobacco users Mean age=28.2 Cannabis and tobacco use of at least 25 days/month for at least 6 months	MWC	Sleep difficulty was reported as a severe withdrawal symptom on the 5th day of abstinence after 9 days of normal use. More severe sleep difficulty during abstinence from both cannabis and tobacco than without abstinence
2009	Wong et al.71	292 boys and 94 girls Ages 3–5 years at baseline and 15–17 at last follow-up	CBC	Sleep problems at ages 3–8 predicted cannabis use onset later in life
2010	Vorspan et al.27	43 cannabis dependent outpatients with a history of opioid dependence and 56 cannabis-dependent outpatients without a history of opioid dependence	In-house withdrawal symptom questionnaire	79.1% of cannabis users with a history of opioid use reported sleep disturbance vs. 53.6% of cannabis users without a history of opioid use
2011	Tringale and Jensen29	147 adult cannabis users divided into two cohorts: those who reported sleep difficulties (n=116) and those who did not report sleep difficulties (n=31) Age range 18–70 Cannabis users were recruited from cannabis clinics	Self-report measures of sleep latency and sleep difficulties	Both cohorts reported increase in sleep latency after cannabis use
2012	McClure et al.44	20 heavy cannabis using adults Mean age=29 Mean cannabis use was 4 times/day for 14 years	PSQI	Cannabis use quantity and duration were positively correlated with reduced sleep quality Years of cannabis use was positively associated with sleep disturbance
2012	Pasch et al.31	723 adolescents Age range 10.8–17.7 years	Self-report measures of sleep duration, sleep patterns	Lower levels of sleep at the baseline predicted greater cannabis use at 2-year follow-up cannabis use at baseline was associated with an increase in weekend oversleeping and total sleep time at 2-year follow-up
2013	Babson et al.47	102 cannabis-dependent military veterans Mean age=50.79 Eligibility was based on meeting the DSM-V criteria for cannabis use disorder	PSQI	Poor sleep quality was related to less treatment success
2013	Babson et al.48	55 cannabis-dependent military veterans Mean age=51.15 Eligibility was based on meeting the DSM-V criteria for cannabis use disorder	PSQI	Poor sleep quality had a greater likelihood of relapse within the first 2 days of a 7-day period of abstinence
2014	Bonn-Miller et al.62	170 adults Mean age=41 Cannabis use was medical	MMQ	Higher PTSD scores were associated with greater frequency of cannabis use and a greater likelihood to use cannabis to promote sleep
2015	Troxel et al.30	2,539 adolescents Mean age=15.54	PHQ and Somatic Symptom Severity Scale	Past month substance use (alcohol and cannabis) was associated with a shorter total sleep time and trouble sleeping
2016	Metrik et al.63	301 veterans with symptoms of PTSD or major depressive disorder Mean age=33.4 Cannabis use was assessed via self-report of use for the 6 months before the session	MPS and MMQ	Motivation for sleep improvements from cannabis use mediated the effects of PTSD and major depressive disorder on cannabis use disorder
2016	Mike et al.68	Mothers and 140 boys interviewed for drug history Age 11 at baseline and 20–22 at follow-up	Child Sleep Questionnaire	Lower sleep quality was associated with earlier cannabis intoxication and repeated use
2016	Conroy et al.66	98 adults categorized as nondaily users (n=29), nonusers (n=20), and daily users (n=49) Mean age=22.3 Cannabis use was assessed via self-reported use over the past 4 weeks before the session	PSQI, ISI, ESS, and MEQ	Daily users (38.8%) had lower PSQI than nondaily users (10.3%) and nonusers (20%).
2017	Cranford et al.58	801 medical cannabis patients with chronic pain Mean age=45.8 Participants were recruited from medical cannabis clinics	JSPQ	59% reported sleep problems related to pain a minimum of 15 times in the past month. Cannabis was reported as helpful for improving sleep. Sleep-related cannabis side effects were present in 35% of respondents
2017	Piper et al.59	1,513 members of dispensaries in New England Mean age=48 Medical and recreational cannabis users were included	In-house self-report survey	The majority of respondents reported using less opioids, sleep medication, anti-anxiety medication, and migraine medication after starting to use medical cannabis
2018	Lim and Thames64	107 HIV+ or HIV− individuals (65 HIV+, 42 HIV−) Mean age=55.12 Cannabis was assessed via self-reported use in the past 30 days	SATED sleep questionnaire	Cannabis use was negatively associated with sleep health HIV− individuals but not HIV+ individuals
2019	Bachhuber et al.61	1,000 adult medical cannabis users Age range 21–65+ Users obtained their cannabis from legal dispensaries, but medical users were excluded	In-house self-report survey	74% of respondents reported using cannabis to promote sleep. 83% of those respondents reported that cannabis was very helpful for sleep, and 87% stopped or reduced their use of over-the-counter medication for sleep
2019	Altman et al.33	311 adult participants with sleep-related problems and cannabis use Mean age=42.39 Participants gave self-reports of cannabis endorsement and monthly use	PSQI	Participants expected cannabis use to decrease their self-reported sleep problems. Expectancies about cannabis influence on sleep problems negatively covaried with cannabis-related problems
2019	Winiger et al.28	1,656 adult twins Mean age=25.79	Jessor Health Questionnaire (sleep duration)	Cannabis use earlier in life was associated with shorter sleep duration in adulthood. There was a significant genetic correlation between the onset of cannabis use and weekend sleep duration
2020	Winiger et al.65	11,875 children Age range 9–10 Prenatal cannabis use was assessed via self-reports of the mother both before and after knowing about the pregnancy	The Sleep Disturbance Scale for Children	Prenatal cannabis use was associated with an increased sleep disorder score, disorders of initiating/maintaining sleep, disorders of arousal/waking, and drowsiness
2021	Winiger et al.28	152 participants Mean age=31.45	PSQI	Cannabis use was associated with expected improvement in sleep from cannabis. Increases frequency of cannabis use was associated with worse subjective sleep quality despite expectations
B. PSG
1974	Barratt et al.34	12 male cannabis users and 4 placebo. Age range 21–26 Cannabis was self-administered via smoking	Polysomnography	Slow-wave sleep initially increased for the cannabis users throughout the first 4 days, but decreased below baseline levels from the 8th day onward
1975	Feinberg et al.35	7 male cannabis users Mean age=25 Cannabis was administered orally (30 mg THC)	EEG	REMs during sleep and REM sleep were reduced after the administration of THC. Stage 4 sleep significantly decreased during a withdrawal period
2004	Nicholson et al.39	8 adult cannabis users Mean age=21.8 Cannabis was administered via oromucosal spray	Polysomnography, self-reports of sleep and alertness	Higher doses of THC (15 mg) were associated with decreased stage 3 sleep, impaired memory, reduced sleep latency, changes in mood and increased sleepiness. CBD doses (15 mg) seemingly counteracted the sedative activity of THC
2008	Bolla et al.41	17 heavy cannabis users and 14 drug-free controls Control mean age=21.7 Cannabis users mean age=20.6 Cannabis users reported >2 years of use and >5 times/week over the past 3 months	Polysomnography, subjective ratings of sleep satisfaction	Cannabis users showed lower total sleep time and less slow-wave sleep than controls over two nights of abstinence
2009	Cohen-Zion et al.57	29 cannabis and alcohol using adolescents and 20 healthy controls Control mean age=18.4 Cannabis users mean age=18.2 Cannabis users reported >200 lifetime uses with <30 lifetime uses of other drugs	Polysomnography	During a 1–2 night abstinence, past month cannabis use was a valid predictor of slow-wave sleep However, during nights 27–28 sleep architecture in adolescence could no longer be predicted by substance use
2010	Bolla et al.50	18 heavy cannabis users in an inpatient unit Mean age=21.1 Cannabis users reported >2 years of use and >5 times/week over the past 3 months	Polysomnography, subjective ratings of sleep quality	During abstinence, total sleep time, sleep efficiency, and REM decreased Wake after sleep onset and periodic limb movements increased
2017	Pacek et al.36	87 cannabis users in an outpatient unit Mean age=31.4 Cannabis users reported at least 50 uses in the past 90 days	Polysomnography, PSQI, ISI	36.8% met the criteria for subthreshold insomnia, and 25.3% for clinical insomnia Only 19.5% received 7–9 h of sleep 31% had sleep-onset latencies of >30 min

CBC, Child Behavioral Checklist; CBD, cannabidiol; DSM-V, Diagnostic and Statistical Manual of Mental Disorders, 5th Edition; EEG, electroencephalography; ESS, Epworth Sleepiness Scale; ISI, Insomnia Severity Index; JSPQ, Jenkins Sleep Problems Questionnaire; MEQ, Morningness Eveningness Questionnaire; MMQ, Marijuana Motives Questionnaire; MPS, Marijuana Problems Scale; MWC, Marijuana Withdrawal Checklist; NRS, Numerical Rating Scale; PHQ, Patient Health Questionnaire; PSG, polysomnography; PSQI, Pittsburgh Sleep Quality Index; PTSD, post traumatic stress disorder; REM, rapid eye movement; SATED, Satisfaction. Alertness. Timing. Efficiency. Duration; THC, tetrahydrocannabinol; VAS, Visual Analogue Scale; WDS, Withdrawal Discomfort Score.

Evaluation of sleep mechanisms typically rely on polysomnography (PSG) studies that include measures of electrical brain oscillations through electroencephalography (EEG), blood oxygen levels, heart rate, and limb movements during sleep (Table 1B). These objective measures provide information on sleep architecture, which is the cyclical pattern of sleep stages, including non-REM and REM sleep, the deepest state of sleep. For example, neural signaling information from EEG such as frontocentral alpha activity has been associated with processes that help maintain sleep. Specifically, increases in frontocentral alpha activity correlate with greater sleep disturbances.²⁴ Another objective measure of sleep is actigraphy that records sleep patterns and circadian rhythms through a noninvasive accelerometer sensor typically worn on the wrist like a watch. These sensors are typically worn for several days to a week to assess the individual's natural sleep/wake cycle. Of interest, a study in healthy adults comparing subjective measures of sleep like the PSQI and sleep diaries with actigraphy found inconsistent findings between the subjective and objective measures of sleep as self-reports of “good” or “poor” sleep classified by the PSQI had no predictive validity for objective sleep quality measured through actigraphy.²⁵ Furthermore, PSQI scores have been shown to be poorly correlated with PSG recordings of the same sample of insomniac patients.²³ This suggests that subjective and objective measures have varying levels of sensitivity to different aspects of sleep. Indeed, Kaplan et al. compared traditional PSG metrics (total sleep time, sleep staging [and times spent in each sleep stage], arousals and awakenings, leg movements, breathing, and apneic episodes) with self-reports of sleep quality within subjects and observed very little predictive validity for subjective sleep quality.²⁶ The likely explanation for such discrepancies in the literature is that the biological recovery and restorative functions of sleep are not reflective of the subjective perception of a satiating night of sleep, although the specific variables that explain the difference between perceived sleep and objective sleep are not yet understood. Thus, future studies should carefully evaluate the most appropriate indicator for the sleep construct of interest.

Given the influence of the ECS on sleep–wake processes, the significance of the relationship between cannabis and sleep is not easily refutable. Although previous reviews have summarized the known associations between cannabis and sleep architecture and sleep quality in humans, particularly in those with sleep disorders, this review will present the current literature on the association between cannabis and sleep toward the goal of (1) understanding when and how these effects manifest and (2) identifying the factors that modulate them. We will also describe the existing human literature from the perspective of chronology as the temporal relationship between sleep disruptions and cannabis use is unclear.

Methods

We conducted a review of the literature on cannabis and sleep in humans through a search of PubMed using the following key words: “cannabis,” “marijuana,” and “THC” as well as “sleep,” “sleep disruption” and “sleep impairment.” Thirty-one studies were identified. Nonhuman literature was excluded and only studies that specifically measured sleep and cannabis were included.

Evidence for Disrupted Sleep Related to Cannabis Use

Studies using self-report measures

The literature regarding how cannabis influences sleep in humans is conflicting. There is evidence from self-report measures through questionnaires or interviews that sustained heavy cannabis use shows disruptions in sleep outcomes, whereas others report improvement in sleep. Vorspan et al.²⁷ found that in 43 treatment-seeking cannabis users, 53.6% reported experiencing sleep disruptions. Sleep quality assessed through the PSQI has been shown to be significantly lower for individuals who endorse current cannabis use relative to those who do not endorse cannabis use.²⁸ In addition, in a sample of 166 cannabis-using individuals, 70% reported sleep difficulties characterized as longer sleep-onset latency and poor perceived sleep quality compared with cannabis users from the sample who did not report sleep difficulties.²⁹ Of interest, the reported sleep problems in the cannabis users were present regardless of history of sleep difficulty, suggesting that cannabis pre-dates the onset of sleep impairment. In a sample of 2,539 adolescents, even when controlling for known risk factors such as parental education or mental health symptoms that may covary with race/ethnicity, increased odds of cannabis use in the past month was associated with later bedtimes and shorter total sleep time.³⁰ Conversely, a longitudinal study of 723 adolescents found that higher cannabis use at the baseline visit was associated with an increase in oversleeping on the weekends, as well as an increase in total sleep time at the follow-up visit.³¹ These conflicting results highlight the discrepancies in the literature regarding how cannabis impacts sleep.

In terms of long-term effects of adolescent cannabis use on sleep, findings are consistent with the larger literature illustrating detrimental effects of cannabis use on brain development impacting function in later life. Cannabis use in adolescence appears to have detrimental effects on sleep that are observable in adulthood. For instance, a study in 1,656 adult twins found that earlier onset of cannabis use (assessed retrospectively) was associated with shorter adult sleep duration compared with later cannabis use in their biological twin.³² The study also observed a genetic association between the onset of cannabis use and weekend sleep in adulthood, which may suggest that the severity of sleep disruptions resulting from cannabis use is modulated by genetic effects. Taken together, these findings suggest a gene-by-environment interaction on dysregulated sleep whereby sleep disruption is triggered by cannabis use in those who are at risk for sleep impairment.

Using subjective measures of sleep, the existing literature shows that cannabis use and sleep problems have a dose–response relationship with drug expectancy. For example, frequency of cannabis use in the past 2 weeks were valid predictors of the expectations of using cannabis as a sleep aid,²⁸ suggesting that continued or increased cannabis use may be perpetuated by the belief of a better sleep outcome from cannabis. Furthermore, a study by Altman et al. found that positive expectancies of cannabis use as a remedy for sleep problems was negatively correlated with cannabis use problems.³³ This suggests that positive beliefs about cannabis use may modulate the perceived relief from sleep disruptions and cannabis use problems.

Evidence from PSG studies

In addition to self-reported measures, alterations in sleep from cannabis use have also been observed through objective measures, such as PSG. The literature on the effects of cannabis on sleep in studies that incorporate PSG also has conflicting results. For example, Barratt et al. tested 12 cannabis users daily over a period of 10 days and found that slow-wave sleep increased in the first 4 days of marijuana use but decreased below baseline levels by the 8th day.³⁴ These findings suggest that although there may be initial sleep benefits from cannabis use, the benefits dissipate quickly. In addition to slow-wave sleep, cannabis was found to decrease REM sleep in four participants after 70 mg/day of THC for 2 weeks.³⁵

Pacek et al. used both PSG and self-report sleep measures such as the PSQI and the ISI to assess sleep quality in heavy cannabis users and found that only 19.5% of the sample had 7–9 h of sleep and 31% had sleep-onset latencies of >30 min.³⁶ Using both the objective and subjective measures of sleep, most participants met the criteria for subthreshold or clinical insomnia.³⁶

Finally, the literature also suggests that many of the features of sleep disruption in chronic marijuana users overlap with medical sleep disorders such as decreases in REM and slow-wave sleep as well as increased sleep-onset latency.³⁷ These observations imply that sleep disruptions in cannabis users are at a level of severity akin to the symptoms of diagnosable sleep disorders such as insomnia. Thus, disentangling the contributions of cannabis from a possible underlying sleep disorder on sleep disturbances is challenging.

Cannabis Use Factors That Are Associated with Sleep Disruption

Dose

Cannabinoid concentrations or dose has been shown to directly modulate sleep. Dose levels can be described as micro dose <1 mg, low dose 1–5 mg, standard dose 5–10 mg, and high dose >10 mg.³⁸ High doses of THC have been associated with changes in sleep architecture such as reduced REM and reduced stage 3 sleep as well as subjective changes such as increased sleepiness.^35,39 High doses of THC administered throughout the day for 7 days to chronic cannabis users were associated with more daytime sleepiness.⁴⁰ Of interest, subjects also reported less difficulty falling asleep and had shorter sleep latencies, suggesting that heavy cannabis users may build a tolerance to the adverse consequences of daily high-dose THC consumption. Despite some conflicting evidence, dose does seem to play an important role in determining the effects of THC on sleep. Although heavy cannabis use is usually associated with adverse sleep consequences, low doses of THC can have a hypnotic effect by increasing slow-wave sleep and decreasing sleep-onset latency.^37,41 The dose–response relationship between cannabis and sleep outcomes is nuanced and not sufficiently understood as of the state of this review.

Of interest, the effects of CBD are opposite of THC's effects on sleep. For instance, higher concentrations of CBD produce beneficial effects on sleep such as longer total sleep time.^42,43 Some animal models indicate that sleep impairments from cannabis may be mitigated by higher CBD content. In male rats, CBD administration at 2.5, 10, and 40 mg/kg showed greatest total sleep time and slow-wave sleep in the 40 mg/kg group.⁴² This finding has been translated in humans in a clinical trial that showed that high doses of CBD were associated with an increase in total sleep time and a decrease in nighttime wake events compared with groups that received lower doses.⁴³

Quantity of use may also be important for determining if individuals will experience sleep disruptions.^37,44 Years of cannabis use is positively correlated with self-reported sleep disturbances.⁴⁴ Taking the literature on sleep disruption during periods of abstinence, we can infer chronology to some degree that heavy cannabis use sustained at a high frequency may produce sleep disruptions upon cessation, or at least exacerbate them if they existed before cannabis use. This evidence of a dose-response relationship between cannabis and sleep provides some support for the directionality of the effect of cannabis on sleep and is further discussed in the section hereunder. It is important to consider the potential that cannabis may simply be exacerbating underlying sleep impairments.

Cannabis withdrawal

Disruption of sleep during periods of abstinence is considered a symptom related to the phenomenon of withdrawal from cannabis. Sleep deficits in cannabis users are most inimical during periods of abstinence as withdrawal symptoms increase the risk for relapse. Vandrey et al. demonstrated that sleep difficulty was reported as the most severe withdrawal symptom after 5 days of abstinence from cannabis use.⁴⁵ In an administration study where young adult participants were given 5–8 days of marijuana (10 mg, THC), significant increases in sleep disturbances were reported following one night of placebo.⁴⁶ Another study reported a similar rate of prevalence of sleep disturbances where 53.6% of 56 cannabis users reported poor sleep as a withdrawal symptom.²⁷ Two studies of military veterans with histories of cannabis use found that participants with higher scores on the PSQI (i.e., worse sleep) had less abstinence success during periods of cessation and were more prone to relapse.^47,48 In a study evaluating withdrawal symptoms in cannabis users and tobacco users, sleep disturbances were reported with a higher prevalence in cannabis users compared with tobacco users.⁴⁹

PSG studies substantiate the self-report data with observations of alterations in sleep architecture as a function of cannabis withdrawal. When PSG was used to compare heavy cannabis users to controls over a two-night period, the cannabis users showed significantly less slow-wave sleep than the controls.⁴¹ This finding was replicated in 18 treatment-seeking cannabis users where a reduction in REM and total sleep time during a 13-day period of abstinence were observed.⁵⁰ These findings indicate that prolonged heavy use of cannabis is associated with suppression of slow-wave sleep during withdrawal.

Adolescent use

Adolescents' high risk for the effects of cannabis on sleep is in line with the well-established notion that cannabis use during adolescence leads to more adverse health and psychosocial outcomes.^51,52 This period of neurodevelopment is critical in establishing brain network connectivity and is highly vulnerable to the impacts of cannabis, and, in turn, the potential disruptions of the sleep cycle. Indeed, variability in sleep duration in noncannabis-using adolescents was associated with reductions in the integrity of the brain's white matter tracts measured through MRI diffusion tensor imaging.⁵³ This reduction was noted in the cingulum, which connects areas enriched with endocannabinoids including the anterior cingulate cortex to the amygdala, nucleus accumbens, the thalamus, and various frontal structures.^54,55 The potential sleep disruption from cannabis use may be significantly harmful to the adolescent brain in its own respect. Cognitive functions like verbal creativity and abstract thinking are impaired in children ages 10–14 after a restriction of just one night of sleep.⁵⁶

Similar to adult cannabis users, the current literature on sleep disruption in adolescent cannabis users is unclear. Pasch et al. found in a sample of 723 adolescents that self-reported cannabis use at baseline was a predictor of an increase in weekend sleep 2 years later.³¹ This observation may imply that cannabis use early in adolescence is associated with increasingly altered sleep patterns in adolescents. Similar to adults, sleep disruption is the most prominent withdrawal symptom after heavy use in the adolescent cannabis-using population. One study recorded PSG data from 29 heavy cannabis-using adolescents and 20 controls during a period of 28 days of abstinence from cannabis and found that slow-wave sleep percentage could be predicted by past month cannabis use compared with noncannabis-using controls.⁵⁷ However, after 28 days of abstinence, cannabis user in an adolescent population could no longer predict changes in sleep architecture,⁵⁷ which may suggest that adolescents are more resilient than adults to the long-term effects of sleep disruption from cannabis. This may also suggest that even in frequent cannabis users the adverse effects of cannabis on sleep may be mitigated within a relatively brief period of abstinence and that long-term effects are minimal. It is also important to note that specificity of these effects in adolescents is difficult to ascertain given significant polysubstance use in this age group.³¹ Moreover, these studies and many others record data from cannabis users that smoke as a primary route of administration, leaving the many alternatives of THC consumption unaddressed. Although the literature regarding how cannabis influences sleep in adolescents is sparse, there is evidence to support the notion that adolescents experience sleep disruption during abstinence similar to adult populations. Furthermore, developing disrupted sleep patterns is uniquely pernicious during adolescence because of ongoing brain development.

Medicinal use

The effects of cannabis for medicinal purposes in clinical populations may differ substantially from those in healthy nonclinical cannabis users particularly in terms of perceived sleep benefits. Cannabis use has been shown to be an effective sleep aid for patients with chronic pain.^58–60 One study that observed patients who used cannabis as a substitute for pain medication for chronic pain found that 84% reported it to be extremely helpful for promoting sleep, with 87% of that group reporting reducing or stopping their intake of over-the-counter medication.⁶¹ Furthermore, those with higher scores of the severity of post traumatic stress disorder (PTSD) symptoms, measured by the PTSD Checklist-Civilian Version, relative to lower scores in PTSD have been found to be more likely to endorse cannabis to improve their sleep.^62,63 Similar associations have also been reported in those with HIV. Specifically, self-reports of sleep health were found to be negatively associated with cannabis use in HIV− subjects, but not in HIV+ subjects.⁶⁴ Taken together, these studies demonstrate that the positive effects of cannabis on sleep are more pronounced in clinical populations. However, the literature is still growing, and replication of these findings is needed.

Are Sleep Deficits a Precursor to or Consequence of Cannabis Use?

To date, our understanding of the chronology of the relationship between cannabis and sleep is limited given that prospective, longitudinal studies on this association are sparse. Furthermore, these types of studies are critical for understanding the chronological relationship between sleep disruptions and cannabis use. A compelling argument that cannabis use may precede sleep impairment comes from a study by Winiger et al. who found that prenatal cannabis exposure was associated with long-term effects on sleep such as problems initiating or maintaining sleep.⁶⁵ Otherwise, as noted previously, the dose–response relationship between cannabis and sleep may be indicative of a causal relationship whereby cannabis leads to sleep deficits. Higher doses of THC and CBD were associated with a decrease in slow-wave sleep and reduced sleep latency compared with lower doses.³⁹ This dose–response relationship is also illustrated in a study where severity of insomnia using the Insomnia Severity Index was greater among daily cannabis users compared with both infrequent users and non-users.⁶⁶ In a sample of 20 inpatient cannabis users, greater volume and duration of cannabis use were correlated with greater reduction in sleep quality.⁴⁴ In sum, high doses of THC sustained for prolonged durations are associated with disrupted sleep outcomes. In alcohol use, such dose–response relationship between alcohol and sleep was attributed to reductions in extracellular levels of adenosine after chronic alcohol use.⁶⁷ Future studies should determine whether altered adenosine may also mediate the relationship between cannabis and sleep.

Conversely, sleep difficulties may be a predisposing factor that contributes to cannabis use. For instance, studies suggest that cannabis users may be self-medicating for sleep problems. Mike et al. showed that low quality of sleep was a valid predictor for cannabis intoxication and escalation of use in a sample of boys 20–22 years of age with reports of sleep duration and quality at age 11 given by the mothers.⁶⁸ A longitudinal study of 4,494 adolescents found that those with symptoms of insomnia were more likely to use cannabis than those without symptoms of insomnia.⁶⁹ There is compelling evidence that poor sleep habits in adolescence, such as the likelihood of getting <7 h of sleep, lead to substance use.⁷⁰ Finally, sleep problems at 3–8 years of age were associated with an increase in the probability of cannabis (and alcohol) use during adolescence.⁷¹ The authors postulate the importance of identifying markers that exist at such an early life stage that can predict adolescent substance. Moreover, after accounting for race, socioeconomic status, internalization of problems, and other potential moderating factors, reduced duration and quality of sleep were associated with earlier cannabis use in adolescents.⁶⁸ In other words, this study also demonstrates that poor sleep is a significant predictor of future cannabis use in adolescents. However, neither of these studies identify how sleep problems develop over time and what variables contribute to cannabis use or other substance use. The literature on sleep disruption before substance use seems to substantiate the claim that sleep problems may be a motivation for individuals to initiate cannabis use.

Discussion

There is sufficient evidence to suggest that cannabis use and sleep influence each other, which is similar to that found in other substances of abuse.^{31,45,50,72,73} One of the most well-supported facets of cannabis use and sleep is that poor sleep is a hallmark symptom of cannabis withdrawal. In addition, cannabis users often form a reliance on cannabis for sleep aid, reaping the initial benefits of cannabis use such as reduced sleep-onset latency. However, the potential benefits of cannabis use on sleep dissipate with time as users develop tolerance to the effects. Heavy sustained cannabis use is associated with worse self-reports of sleep quality, reduced slow-wave sleep, and potentially reduced REM sleep. These observations seem to be exacerbated during periods of abstinence, which may contribute toward relapse. Conversely, poor sleep, specifically in younger populations, has been shown to be an index of future cannabis use. The temporal relationship of cannabis and sleep disruption is unclear, but the evidence suggests that poor sleep may be a consequence of cannabis use and a predictor for unsuccessful attempts to quit.

There is emergent literature regarding the effects of cannabis use on sleep, which complicates the issue given that healthy adolescents lose as much as 120 min of sleep on weeknights leading to altered sleep patterns from week days to weekends.⁷⁴ Adding to this concern, during adolescence, the brain undergoes crucial stages of growth and sleep duration has been associated with changes in volume of several brain areas.¹⁶ It is unclear if sleep disturbances arise after cannabis use in adolescents or, rather, if sleep disturbances exist as a precursor that motivates adolescents toward cannabis use. Future studies should apply prospective, longitudinal designs that utilize comprehensive measures and technologies that allow recording of data in naturalistic settings over a duration of period. This would provide information on how habitual sleep habits ebb and flow throughout different stages of cannabis use, painting the clearest picture of the chronology of effects during extended periods of cannabis use and abstinence.

A limitation of most of the studies on cannabis and sleep is a reliance on self-report measures to quantify sleep quality. Many of the modalities that are critical to assessing sleep are recorded through self-reports through questionnaires, diaries, or surveys. This may open the door for future research to develop methods for attaining the same information through objective measures. Corroborating subjective assessments of sleep problems with objective measures may have the potential to better identify reliable biomarkers of sleep deficits associated with cannabis use. Such biomarkers, for instance, could help identify users who are at risk for relapse. Furthermore, advancements in technologies such wearable devices that are capable of recording polysomnographic data could be a novel self-therapeutic platform.

Although there is growing interest in how cannabis impacts sleep in the short term, there is a surprising lack of literature that investigates the long-term indirect effects of cannabis users with impaired sleep. Cannabis users with sleep-related problems may be at a higher risk for developing cognitive deficiencies or problems in daily function, but there is not enough evidence to reach any conclusions as of the current state of the literature. One study using structural equation modeling found that the negative association between substance use such as alcohol/cigarette and academic success may be mediated by daytime sleepiness.⁷² However, this study did not account for cannabis use. This review highlights that younger populations are at risk for developing a proper sleep schedule if they engage in sustained cannabis use. Alterations in sleep patterns and sleep architecture may be specifically inimical to adolescents by disrupting brain development. Furthermore, there are a surprising lack of studies that investigate cannabis and sleep with a focus on additional cognitive assessments. There is growing evidence for how cannabis affects sleep in the short term, but the long-term effects on cognition are not well understood. There has been previous finding that document how detrimental altered sleep can be in healthy populations. Changes in sleep-onset latency, sleep patterns, and sleep duration have all been shown to have cognitive deficits. Cannabis users who experience sleep disturbance often report many of these symptoms, meaning they may be at an increased risk for negative long-term side effects. Future studies should investigate how poor sleep within chronic cannabis users impact cognitive factors such as memory, IQ, or executive functioning.

With cannabis use continuing to be more prevalent, there is a growing need for a thorough investigation of the potential risks and benefits of cannabis use on sleep. There are many factors that influence the outcome cannabis has on sleep. Factors like quantity/duration of use, age, THC content, or pre-existing medical conditions may influence the development and severity of sleep disruption. Focusing on the behavioral and biological traits within these populations will facilitate a nuanced understanding of the circumstances when cannabis may or may not be an effective sleep aid.

Abbreviations Used

CB1

cannabinoid 1

CBC

Child Behavioral Checklist

CBD

cannabidiol

DSM-V

Diagnostic and Statistical Manual of Mental Disorders, 5th Edition

ECS

endocannabinoid system

EEG

electroencephalography

ESS

Epworth Sleepiness Scale

ISI

Insomnia Severity Index

JSPQ

Jenkins Sleep Problems Questionnaire

MEQ

Morningness Eveningness Questionnaire

MMQ

Marijuana Motives Questionnaire

MPS

Marijuana Problems Scale

MWC

Marijuana Withdrawal Checklist

NRS

Numerical Rating Scale

PHQ

Patient Health Questionnaire

PSG

polysomnography

PSQI

Pittsburgh Sleep Quality Index

PTSD

post traumatic stress disorder

REM

rapid eye movement

SATED

Satisfaction. Alertness. Timing. Efficiency. Duration

THC

tetrahydrocannabinol

VAS

Visual Analogue Scale

WDS

Withdrawal Discomfort Score

Author Disclosure Statement

No competing financial interests exist.

Funding Information

F.M.F. holds the Bert Moore Chair and is supported by NIH R01DA042490.

Cite this article as: Edwards D, Filbey FM (2021) Are sweet dreams made of these? Understanding the relationship between sleep and cannabis use, Cannabis and Cannabinoid Research 6:6, 462–473, DOI: 10.1089/can.2020.0174.