Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Nov 1.
Published in final edited form as: Drug Alcohol Depend. 2023 Oct 11;252:110989. doi: 10.1016/j.drugalcdep.2023.110989

Assessing Changes in Sleep Across Four Weeks Among Adolescents Randomized to Incentivized Cannabis Abstinence

Andreas M Baumer a,*, Bridget A Nestor a,b,*, Kevin Potter b,c, Sarah Knoll c, A Eden Evins b,c, Jodi Gilman b,c, Joe Kossowsky a,b,**, Randi M Schuster b,c,**
PMCID: PMC10691527  NIHMSID: NIHMS1939802  PMID: 37839357

Abstract

Background:

Withdrawal from cannabis use is associated with sleep disturbances, often leading to resumption of use. Less is known about the impact of abstinence on sleep in adolescence, a developmental window associated with high rates of sleep disturbance. This study investigated effects of sustained abstinence on self-reported sleep quality and disturbance in adolescents reporting frequent cannabis use.

Methods:

Non-treatment seeking adolescents, recruited from school screening surveys and the community, with frequent cannabis use (MAge=17.8, SDAge=1.7, 47% female, 45% non-white) were randomized to four weeks of biochemically-verified abstinence, motivated via contingency management (CB-Abst, n=53), or monitoring without an abstinence requirement (CB-Mon, n=63). A mixed-effects model was used to predict change in Pittsburgh Sleep Quality Index (PSQI) scores.

Results:

Participants in CB-Abst reported higher overall PSQI scores than those in CB-Mon (M=1.06, p=0.01) indicating worse sleep during the four-week trial. Sleep disruptions in CB-Abst increased during Week 1 of abstinence (d=0.34, p=0.04), decreased during Week 2 (d=0.36, p=0.04), and remained constant for the rest of the trial. At Week 4, sleep was comparable to baseline levels for those in CB-Abst (p=0.87). Withdrawal-associated sleep disruption in the CB-Abst group was circumscribed to increases in sleep latency (b=0.35; p=0.05).

Conclusions:

Cannabis abstinence in adolescents was associated with transient delayed onset of sleep initiation during the first week of abstinence. Findings highlight withdrawal-associated changes in sleep latency as an intervention target for supporting adolescents attempting abstinence. Future research should use objective measures of sleep and focus on elucidating mechanisms underlying sleep disturbances with cannabis use and withdrawal.

Keywords: cannabis use, withdrawal, sleep, adolescence

1. Introduction

After alcohol, cannabis is the most commonly used psychotropic substance by adolescents in the United States, with approximately 38% and 6% of 12th graders reporting lifetime and daily use, respectively, in 2022 (Miech et al., 2023). Adolescence is a period marked by high rates of sleep disturbance (Kansagra, 2020), and improvement in sleep is a commonly cited reason for cannabis use in this population (Goodhines et al., 2022). Acute withdrawal from cannabis use, however, is also associated with sleep disturbances (Angarita et al., 2016), often leading to resumption of use (Babson & Bonn-Miller, 2014). Less is known, though, about the specific impact of sustained abstinence on sleep changes in adolescence, which could inform interventions aimed at supporting adolescents through abstinence attempts.

Emerging evidence shows that acute and chronic exposure to Δ-9-tetrahydrocannabinol (THC) have putative effects on sleep in adolescents (Cohen-Zion et al., 2009). Animal and human studies indicate that acute THC exposure is associated with sleep-promoting effects, such as increased sleep consolidation, decreased sleep onset latency, increased total sleep time, and decreased waking after sleep onset (Angarita et al., 2016; Kaul et al., 2021). These effects are likely mediated by activation of cannabinoid (CB1) receptors, which play an important role in regulating the sleep-wake cycle (Kesner & Lovinger, 2020). However, repeated exposure to cannabis and THC, in particular, has repeatedly been shown to produce tolerance to the sleep enhancing drug effects (Barratt et al., 1974; Freemon, 1982; Gorelick et al., 2013; Halikas et al., 1985; Karacan et al., 1976; Pranikoff et al., 1973), thus requiring increasing dosages to obtain the sleep-promoting action. The need for increased dosages is problematic given evidence showing repeated heavy exposure to cannabis during adolescence is associated with marked anatomical changes in the brain (e.g., altered cerebral cortical development), poor functional outcomes (e.g., future substance dependance), and increased propensity for serious mental health disorders (Di Forti et al., 2019; Gruber et al., 2014; Lisdahl et al., 2014; Quattrone et al., 2021; Schweinsburg et al., 2008; Stockman, 2009). Chronic THC exposure has been associated with disruptions to sleep and sleep architecture, including increased sleep onset latency, increased wake after sleep onset, and reduced total sleep time (Pacek et al., 2017). While effects on rapid eye movement (REM) are non-uniform, numerous studies report decreased slow wave sleep among those with chronic cannabis use (Adams & Barratt, 1975; Barratt & Adams, 1973; Feinberg et al., 1976; Pranikoff et al., 1973), which have been associated with deficits in learning, memory, and cognitive performance (Dijk, 2010).

Acute cannabis withdrawal is also known to exacerbate sleep disturbances (Angarita et al., 2016; Gates et al., 2016), including trouble falling asleep, waking during the night, distressing dreams, and reductions in sleep time (Gates et al., 2016), which may represent a salient trigger for relapse in those attempting abstinence (Babson & Bonn-Miller, 2014). However, most studies investigating the effects of cannabis withdrawal on sleep have used measures that are not sleep-specific (Allsop et al., 2011) or have employed only retrospective (Vandrey et al., 2005) or observational (Milin et al., 2008) assessments of sleep changes.

Advancing our understanding of the effects of acute cannabis withdrawal on sleep for adolescents, one study compared sleep in adolescents undergoing incentivized abstinence and in a control sample of non-using adolescents over the course of 4 weeks(Jacobus et al., 2017; Sullivan et al., 2022) and reported differences in sleep quality and sleep disturbance that resolved after 4 weeks of abstinence, though abstinent adolescents continued to report less overall sleep compared to non-using adolescents. Methodologically, this study design lacked randomization, did not compare sleep to a sample of adolescents who continued cannabis use, and assessed an overall sample that reported subclinical sleep throughout the duration of the study, thus limiting broad generalizability of findings. In a similar investigation (e.g., Sullivan et al., 2022), researchers again compared sleep in a sample of adolescents during monitored abstinence and in a control (non-using) group. Findings from this study did not reveal a linear decrease in sleep disturbance but rather showed a quadratic trend of initial increase in sleep disturbance for abstinent adolescents that later resolved after 3 weeks of abstinence. This study was again limited methodologically by lack of randomization, lack of comparison to adolescents who continued cannabis use, as well as use of only two items to measure changes in sleep longitudinally.

To build upon prior work, more rigorous methodology is needed in order to clarify the potential persistent effect of cannabis on sleep dimensions during adolescence. Specifically, randomized controlled trials of extended cannabis abstinence, compared to continued use, with longitudinal assessment using validated measures of sleep are necessary. Clarifying how sleep functioning may change across the first month of abstinence in adolescents, beyond periods of acute withdrawal, will yield clinically useful information that can guide treatment planning efforts. The current study aimed to understand the effects of four weeks of biochemically-verified, incentivized cannabis abstinence compared to continued use on various metrics of self-reported sleep health in adolescents who use cannabis at least weekly. Based on findings from previous studies (e.g., Jacobus et al., 2017; Sullivan et al., 2022), we hypothesized that there would be a negative impact of cannabis withdrawal on overall sleep, but that such withdrawal-induced sleep disturbance would be transient and resolve by the end of the 4 weeks to be better than pre-abstinence levels. Due to mixed findings and methodological variability in prior investigations, however, we did not set forth specific hypotheses with respect to changes within various components of sleep quality or disturbance, and instead approached these aims as exploratory. Furthermore, we also explored how sleep disturbance in those with sustained abstinence and continued use compared to non-using adolescents to estimate the degree of disturbance at baseline and the extent of normalization once use has been discontinued.

2. Methods

This study occurred as part of a larger parent project aiming to evaluate the impact of cannabis abstinence on cognition (NCT03276221). All study procedures were approved by the Mass General Brigham Human Subjects Committee.

2.1. Participants and Recruitment

Participants were recruited via school screening surveys at middle and high schools in Greater Boston, community advertisements, and peer referrals. Recruitment was conducted from August 2017 to November 2022 and included 13- to 19-year-olds. In addition to age, general inclusion criteria were English fluency, no history of severe developmental delays, and being medically healthy. Additional eligibility criteria for enrolled adolescents with cannabis use included at least weekly cannabis use, cannabis use in the week prior to the baseline visit, and willingness to abstain from cannabis use for 30 days. Participants with cannabis use were also not actively seeking treatment for their cannabis use and denied having an intent to discontinue cannabis use in the next 30 days. Eligible adolescents without cannabis use reported fewer than five instances of cannabis use in their lifetime, no cannabis use in the past year, and no cannabis use before age 16.

2.2. Procedure

Informed written consent for participants 18-years or older was obtained prior to beginning study procedures. Parental consent and participant assent were obtained for participants under the age of 18. Following a baseline assessment, cannabis-using participants were randomized to four weeks of either incentivized cannabis abstinence (CB-Abst) or cannabis monitoring with no abstinence requirement (CB-Mon). For CB-Abst, a contingency management plan was implemented such that participants were provided with escalating monetary incentives contingent upon continuous, biochemically-verified cannabis abstinence throughout the four weeks. Biochemical verification of abstinence occurred six times: 2-days, 3-days, 1-week, 2-weeks, 3-weeks, and 4-weeks after baseline. Each urinary assay included several checks for adulteration, including temperature checks, detection of the presence of oxidizing agents, and measures of dilution, including pH, specific gravity, and creatinine. No sample collected was suggestive of tampering. CB-Mon participants were provided with incentives for completion of assessments without the requirement of abstinence. Randomization was stratified by sex assigned at birth (male vs. female), age, and frequency of cannabis use. Participants not using cannabis (Non-Using) completed visits at baseline and 1-week, 2-weeks, 3-weeks, and 4-weeks after baseline. As data from the current study are drawn from a larger parent project, details regarding study design and procedures are reported in previously published investigations from the parent project (Schuster et al., 2016; Schuster et al., 2020).

2.3. Measures

The Pittsburgh Sleep Quality Index (PSQI) (Buysse et al., 1989) is a 19-item self-report questionnaire, which assesses seven facets of sleep quality and disturbance, including subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbance, use of sleeping medication, and daytime dysfunction. Each item is rated on a scale from 0 to 3, with higher values indicating greater sleep disturbance and sum scores greater than 5 considered poor sleep. The PSQI has adequate reliability and validity in adolescents (Raniti et al., 2018). Although validated as a 4-week assessment, the PSQI has been used effectively in other studies as a weekly measure (e.g., Jacobus et al., 2017). All participants completed the PSQI at baseline and at the four weekly assessments after baseline (1-week, 2-weeks, 3-weeks, and 4-weeks). Participants were asked to report on their sleep quality over the past four weeks at baseline and then over the past week at each weekly visit. For sleep latency, efficiency and duration, free-text responses also were recorded, which were preferentially analyzed as they offer greater variance (see procedure used in (Wallace et al., 2022)).

The Mood and Anxiety Symptom Questionnaire Short Form (MASQ-SF) (Watson & Clark, 1991) is a 62-item self-report questionnaire assessing mood and anxiety symptoms with four subscales: general distress anxious symptoms (GDA), anxious arousal (AA), general distress depressive symptoms (GDD), and anhedonic depression (AD). Each item is rated on a 1 to 5 scale, with higher values indicating greater symptom frequency. In the current study, the GDA and AA subscales were summed to create an anxiety composite score and the GDD and AD subscales were summed up to create a depression composite score. The MASQ-SF has good reliability and validity (Watson et al., 1995). All participants completed the MASQ-SF at baseline and at the four weekly assessments after baseline (1-week, 2-weeks, 3-weeks, and 4-weeks).

The Cannabis Use Disorder Identification Test – Revised (CUDIT-R) (Adamson et al., 2010) is an 8-item questionnaire that assesses cannabis use in terms of frequency, impairment, and failed attempts at quitting. Each item is scored from 0 to 4. Total scores greater than 8 indicate hazardous use and scores greater than 12 indicate a possible cannabis use disorder. This measure has shown good reliability and validity (Adamson et al., 2010). Participants in CB-Abst and CB-Mon completed the CUDIT-R at baseline.

3. Analytic Approach

Participants were included in analyses if they had data available at baseline as well as at any post-randomization time point. If participants in CB-Abst resumed cannabis use during the four-week intervention, their data were included in analyses until abstinence was discontinued. To assess the effects of four weeks of cannabis abstinence on overall self-reported sleep health for the CB-Abst and CB-Mon groups, we fit a mixed-effects model with PSQI total score as a dependent variable and the following predictors: cannabis use group (CB-Abst vs CB-Mon), time (linear; days since beginning of trial), time (quadratic; days since beginning of trial squared), group x linear time interaction, and group x quadratic time interaction. Of note, we included a quadratic term in the models as we hypothesized sleep disturbance would increase during abstinence and then resolve, consistent with patterns observed in prior work (e.g., Sullivan et al., 2022). Covariates included biological sex (male vs female), age, baseline and time-varying MASQ-SF anxiety composite score, baseline and time-varying MASQ-SF depression composite score, and baseline CUDIT-R score. A random PSQI intercept and slope over linear time were added for each participant. A random slope over quadratic time could not be added as it prevented model conversion. We conducted additional analyses using the same model structure for each individual PSQI component scale, by fitting the same model for each subscale of the PSQI in addition to our main sum score model. To further assess changes in sleep over time during abstinence, we conducted consecutive paired t-tests to compare the PSQI scores of participants in CB-Abst at each time point. Results were adjusted for multiple comparisons using the Benjamini-Hochberg method.

To explore whether temporal patterns in self-reported sleep varied in the CB-Abst and CB-Mon groups compared to the Non-Using group, we fit a linear regression model using Generalized Estimating Equations (GEE). We assumed auto-correlations for the time points (baseline and four follow-up visits) followed an independence correlation structure; we note that GEE results are robust to misspecification of this correlation structure (Højsgaard et al., 2006). This model included the same covariates as the mixed-effects model, though did not include the CUDIT-R, as this measure was not administered to Non-Using. Results were adjusted for multiple comparisons using the Benjamini-Hochberg method.

All analyses were conducted in R using the packages lme4 (v1.1.29) (Bates et al., 2014) and geepack (v1.3.9) (Højsgaard et al., 2006; Yan, 2002; Yan & Fine, 2004), and significance values were computed using the lmerTest (v3.1.3) (Kuznetsova et al., 2017) and emmeans (v1.8.5) (Kuznetsova et al., 2017) packages.

4. Results

4.1. Participants

The final sample for primary analyses included 116 participants (nCB-Abst=53; nCB-Mon=63) with a mean age of 18.26 (SD=1.40) years. Table 1 displays key participant characteristics. Cannabis use groups were comparable across all assessed baseline characteristics (see Supplementary Material). The Non-Using group (nNon-Using = 66), which was included in exploratory analyses, was younger than either cannabis using group (AgeNon-Using – CB-Abst: 1.31, p< .001; AgeNon-Using – CB-Mon: 1.26, p<.001) and reported lower MASQ Depression scores (MASQ-DNon-Using – CB-Abst=−8.32, p=0.01; MASQ-DNon-Using – CB-Mon=−7.31, p=0.02) at baseline.

Table 1.

Demographic, cannabis use-related, and clinical characteristics at baseline.

CB-Abst CB-Mon Non-Using
N = 53 N = 63 N = 66
Demographics
 Sex
  Female (%) 25 (47.17%) 29 (46.03%) 38 (57.58%)
 Race
  American Indian or Alaska Native (%) 1 (1.89%) 1 (1.59%) 0 (0%)
  Asian (%) 4 (7.55%) 6 (9.52%) 6 (9.09%)
  Haitian or Black or African American (%) 9 (16.98%) 10 (15.87%) 14 (21.21%)
  More than one race (%) 6 (11.32%) 11 (17.46%) 3 (4.55%)
  Other (%) 2 (3.77%) 3 (4.76%) 5 (7.58%)
  White (%) 31 (58.49%) 32 (50.79%) 38 (57.58%)
 Ethnicity
  Hispanic or Latino (%) 5 (9.43%) 16 (25.40%) 8 (12.12%)
 Agea,b
  Mean (95% CI) 18.30 (17.92 - 18.67) 18.24 (17.88 - 18.60) 16.98 (16.52 - 17.43)
Cannabis Use
 Age at first cannabis use
  Mean (95% CI) 14.96 (14.46 - 15.46) 14.78 (14.35 - 15.20) ---
 Number of days cannabis was consumed in the last week
  Mean (95% CI) 3.67 (2.89 - 4.45) 3. 33 (2.67 - 3.99) ---
 CUDIT-R Sum Score
  Mean (95% CI) 13.45 (12.03 - 14.88) 13.75 (12.45 - 15.04) ---
Clinical Characteristics
 MASQ
  Anxiety sum score
  Mean (95% CI) 42.34 (38.93 - 45.75) 43.27 (39.71 - 46.83) 41.00 (37.98 - 44.02)
  Depression sum scorea,b
  Mean (95% CI) 79.42 (74.37 - 84.46) 78.40 (73.83 - 82.96) 71.09 (66.63 - 75.55)
 PSQI
  Sum Score
  Mean (95% CI) 5.77 (4.79 - 6.76) 5.33 (4.50 - 6.12) 4.91 (4.18 - 5.64)
Open in a new tab

Note. CB-Abst: Cannabis using participants randomized to 4 weeks of incentivized abstinence; CB-Mon: Cannabis using participants randomized to monitoring only; CUDIT-R: Cannabis Use Disorder Identification Test - Revised; MASQ: Mood and Anxiety Symptom Questionnaire; PSQI: Pittsburgh Sleep Quality Inventory.

Pairwise χ2- and t-tests were run for the three groups. Subscript letters denote significant differences (p < .05).

a)

Significant CB-Abst versus Non-Using difference; and

b)

Significant CB-Mon versus Non-Using difference.

Due to small counts, racial identifiers were collapsed to White vs Other-Races for statistical testing. CB-Abst and CB-Mon were statistically comparable across all assessed baseline characteristics.

4.2. Change in Overall Sleep by Randomized Cannabis Group (Main Analysis)

There was a main effect of cannabis use group on PSQI sum score, with CB-Abst reporting on average 1.06 points higher than CB-Mon (p=0.01), reflecting worse sleep across the intervention period. There was not a main effect of time or an interaction between group and time on PSQI sum scores. Consecutive paired t-tests of PSQI sum scores for CB-Abst revealed an increase in sleep difficulties from Baseline to Week 1 (d=0.34, p=0.04) and a decrease from Week 1 to Week 2 (d=0.36, p=0.04) of abstinence. No differences were found in Week 3 or Week 4 compared to Week 2 in CB-Abst (p-values >0.65). No differences emerged comparing baseline to Week 2, Week 3, or Week 4 (p-values > 0.87). All paired t-tests of PSQI sum scores for CB-Mon were non-significant (p-values >0.29). See Figure 1 and Table 2.

Fig. 1.

Fig. 1.

Open in a new tab

Sleep scores of cannabis using participants over trial duration. Note: Mean Pittsburgh Sleep Quality Inventory (PSQI) sum scores including 95% CI. Scores above 5 indicate poor sleep as denoted by the dotted line.

Table 2.

Predictors of the PSQI sum score and the component scale models.

Target PSQI Scale (Dependent Variable)
Predictors Sum Score Sleep Latency Daytime Dysfunction Sleep Quality Sleep Disturbance Sleep Duration Sleep Efficiency Sleep Medication
Intercept 0.61 (0.831) 0.54 (0.27) −0.58 (0.408) 0.08 (0.898) −0.01 (0.984) 5.45 (<0.001) 4.60 (<0.001) 0.49 (0.566)
Sex (Difference to Male) −1.22 (0.006) −1.5 (0.04) −0.25 (0.019) −0.20 (0.027) −0.15 (0.035) 0.09 (0.68) 0.05 (0.003) −0.04 (0.767)
Age −0.04 (0.766) 0.04 (0.15) −0.18 (0.612) 0.01 (0.630) 0.01 (0.624) 0.13 (0.09) < 0.01 (0.84) −0.02 (0.621)
MASQ4 Anxiety 0.07 (<0.001) <0.01 (0.96) 0.02 (<0.001) 0.01 (0.003) 0.01 (<0.001) −0.01 (0.07) < 0.01 (0.21) 0.01 (0.108)
MASQ Depression 0.03 (<0.001) <0.01 (0.05) 0.01 (<0.001) 0.01 (0.022) <0.01 (0.196) < 0.01 (0.72) < 0.01 (0.005) < 0.01 (0.946)
CUDIT-R 0.08 (0.061) 0.01 (0.06) 0.02 (0.023) 0.02 (0.057) 0.01 (0.028) −0.01 (0.55) < 0.01 (0.65) 0.01 (0.407)
Group Effect (Difference to CB-Mon) −1.06 (0.013) −0.04 (0.63) −0.21 (0.045) −0.13 (0.143) −0.08 (0.235) 0.24 (0.68) 0.02 (0.32) −0.21 (0.100)
TimeLinear (CB-Abst) 0.61 (0.125) −0.06 (0.42) 0.08 (0.534) 0.19 (0.059) 0.11 (0.239) −0.23 (0.28) < 0.01 (0.94) −0.02 (0.847)
TimeQuadratic (CB-Abst) −0.84 (0.070) −0.11 (0.12) −0.10 (0.458) −0.25 (0.022) −0.13 (0.213) 0.37 (0.10) < 0.01 (0.99) 0.08 (0.469)
TimeLinear x Group Interaction −0.88 (0.110) −0.28 (0.004) 0.02 (0.903) −0.14 (0.284) −0.09 (0.434) 0.21 (0.43) 0.02 (0.54) −0.01 (0.919)
TimeQuadratic x Group Interaction 0.78 (0.164) 0.27 (0.002) −0.02 (0.911) 0.19 (.153) 0.07 (0.553) −0.34 (0.22) < 0.01 (0.85) −0.02 (0.858)
Open in a new tab

Note: These models include Cannabis Abstinence (CB-Abst) and Cannabis Monitoring (CB-Mon) data, and the reference group is CB-Abst. Results report unstandardized beta coefficients (p-values). Significant results are bolded. Sleep Latency and Sleep Duration are measured in hours, Sleep Efficiency as described by Wallace et al. (2022). Daytime Dysfunction, Sleep Quality, Sleep Disturbance and Sleep Medication are ordinally rated from 0 to 3.

CB-Abst: Abstinence Group; CB-Mon: Monitoring Group; PSQI: Pittsburgh Sleep Quality Inventory; MASQ: Mood and Anxiety Symptom Questionnaire; CUDIT-R: Cannabis Use Disorder Identification Test - Revised;

4.3. Change in Components of Sleep by Randomized Cannabis Group (Exploratory Analysis)

Table 2 displays the estimated effects and p-values for each predictor for each component scale. Higher MASQ Depression composite scores and female sex were the most common predictors of disturbed sleep, both being significant across four individual component scales. Sleep Latency was the only PSQI subscale with significant time x group interactions (Linear: b=−0.28, p=0.004; Quadratic: b=0.27, p=0.002), indicating a stronger rise and decline for CB-Abst than for CB-Mon. There was also a main effect of group, with CB-Abst exhibiting higher Daytime Dysfunction than CB-Mon (b=0.21, p=0.05) across time points, indicating a temporary worsening of sleep latency in CB-Abst compared to CB-Mon. Consecutive paired t-tests show a trend level increase from Baseline to Week 1 (d=0.27, p=0.05) and a decrease from Week 1 to Week 2(d=0.25, p=0.08) within CB-Abst.

4.4. Comparison of Overall Sleep between the Cannabis Groups and Non-Using Group (Exploratory Analysis)

Figure 2 shows the GEE adjusted PSQI group means for each time point as well as the corresponding 95% CI. The Non-Using and CB-Mon groups were comparable across all time points (p-values>0.25). Significant differences were found at Week 1 between the CB-Abst and CB-Mon (b=1.61, p=0.04) as well as the Non-Using group (b=1.38, p=0.04) and at Week 4 between the CB-Abst and the CB-Mon group (b=1.49, p=0.04). The CB-Abst group was comparable to the Non-Using group in Weeks 2 through 4 (b-values<1.12, p-values>0.13).

Fig. 2.

Fig. 2.

Open in a new tab

Adjusted sleep scores of all participants over trial duration. Note: Mean GEE model fitted PSQI1 sum scores including 95% CI. Scores above 5 indicate poor sleep. Significant between group differences were found at Week 1 between Abstinence and Monitoring as well as Abstinence and Non-Using.1PSQI: Pittsburgh Sleep Quality Inventory.

5. Discussion

The current study investigated the effects of four weeks of biochemically-verified, incentivized cannabis abstinence compared to continued use on various metrics of self-reported sleep quality and disturbances in adolescents. Consistent with withdrawal-associated sleep disturbances observed in other studies (Gates et al., 2016; Milin et al., 2008; Vandrey et al., 2005), our main findings indicated initial sleep difficulties for abstinent adolescents. These difficulties peaked during the first week, returned to baseline levels during the second week, and remained at baseline levels for the remainder of the four-week intervention. At the end of the intervention, self-reported sleep in the abstinent group was comparable to that of non-using adolescents.

Additional analysis of individual sleep components revealed that initial sleep challenges during abstinence were driven by changes in sleep latency, instead of a more diffuse set of sleep difficulties across the dimensions of sleep health. This finding suggests that “trouble falling asleep” may be the mechanism driving the rise and subsequent fall we observed in reported sleep issues during cannabis abstinence, which is consistent with findings from prior work (Bolla et al., 2008; Vandrey et al., 2011). Biologically, this initial increase and then resolution of sleep latency challenges may be in line with the acute depletion of CB1 receptor availability and associated initial escalation of withdrawal symptoms, observable several days after cessation and then typically resolved after a month of abstinence (D’Souza et al., 2016).

Our results also revealed that sleep disruptions for adolescents in the monitoring group decreased over the four-week intervention. This finding tempers strong conclusions that sleep improvements in the abstinence group can be exclusively attributed to the diminishing withdrawal effects. Rather, this sleep improvement during a monitoring period may be consistent with other studies that have demonstrated associations between monitoring and symptom reduction. For example, monitoring of cannabis use has been shown to be associated with decreases in use in adolescents (Shrier et al., 2018) and keeping a sleep diary has been shown to be related to decreases in insomnia complaints (Philip et al., 2022).

Contrary to our hypothesis, we could not find evidence that cannabis use either improves or worsens sleep as evidenced by the findings that 1) at baseline both randomized cannabis use groups are comparable to the non-using adolescents and 2) after 4 weeks of abstinence, sleep is comparable to baseline levels. Further exploratory comparisons indicated that non-using adolescents and those who maintained continued use reported very similar sleep scores. This is somewhat unexpected, as many individuals using cannabis believe that their use is associated with sleep-promoting effects; specifically, a recent survey found that 64% of individuals reported consuming cannabis to treat their sleep disorders (Suraev et al., 2023). The absence of an observed difference in sleep outcomes in our sample may instead provide support that the sleep-promoting effects of cannabis diminish with chronic use (Angarita et al., 2016). Non-using adolescents, however, could not be randomized to their group, so we recommend cautious interpretation of this finding as it does not support causal inference.

Taken together, results from the current study suggest that sleep challenges secondary to cannabis abstinence are circumscribed to changes in sleep latency, which arise during the first week and then resolve over a month of sustained abstinence. Clinically, this time-limited and specific sleep difficulty represents an intervention target for sleep hygiene and other CBT-I (Taylor & Pruiksma, 2014) strategies to decrease sleep latency in adolescents, particularly during the initial week of an abstinence attempt. Indeed, a recent intervention demonstrated that CBT tailored to individuals with frequent cannabis use for sleep is effective for improving sleep outcomes as well as reducing cannabis use (Arnedt et al., 2023). Similar intervention efforts employing CBT aiming to limit or eliminate cannabis use in adolescents may maximize their efficiency by focusing on trouble falling asleep during initial days of abstinence, as identified in the current study.

Several limitations of the current study represent avenues for future research. First, the current study used a self-report measure of sleep, which is subject to reporter biases. Future work would be strengthened by use of objective measures. Second, the current study was limited to a four-week period of abstinence and monitoring, limiting the generalizability of our findings longitudinally. We suggest future longitudinal work examine trends in sleep following the end of biochemically verified sustained abstinence, as some research suggests cannabinoids may not be fully cleared from the central nervous system after four weeks of abstinence (Schuster et al., 2020). Longer term follow-up studies with longer periods of abstinence are needed to more clearly determine the possible effects of sustained cannabis abstinence on sleep. Third, adolescents with cannabis use in the current study were heterogeneous in terms of their cannabis use, with some reporting daily or more frequent use and others reporting only weekly use. Whether changes in sleep during abstinence differ based on frequency of cannabis use thus cannot be determined by the current study. Fourth, additional details about adolescents’ reasons and motivations for use (e.g., whether or not their use was for sleep aid) and the chronicity of their use prior to randomization were not investigated in the current study (e.g., whether or not prior use changed in anticipation of enrollment in this research study, etc.). Future work would benefit from clarifying changes in sleep during abstinence for adolescents with different “types” of cannabis use (e.g., instrumental vs. recreational use, etc.). We also do not have specific information about potency of participants’ cannabis use, which research has identified to be variable (ElSohly et al., 2016) and dose-dependently associated with other adverse outcomes of cannabis use (Polkosnik et al., 2021). To address this limitation, we explored the effects of baseline cannabis metabolite concentration (THCCOOH) as the closest available proxy for burden of THC exposure on study outcomes, and found no relationship with change in both withdrawal symptoms (p > 0.70) and self-reported sleep (p > 0.58) over the course of the 4 week trial. Future studies should explore the effects of potency measured more directly on sleep disturbance. Finally, we note that the small sample size in the current study may limit its generalizability to larger, more diverse samples of youth.

Conclusion

Cannabis abstinence in adolescents reporting at least weekly use is associated with sleep disruption, primarily within the first week of abstinence. This pattern of initial worsening and subsequent improvement of sleep may be specifically driven by sleep latency instead of a more diffuse effect across sleep dimensions. Given the known deleterious associations of cannabis use in adolescence (Degenhardt et al., 2013; Lorenzetti et al., 2020; Lynskey & Hall, 2000), a thorough understanding of sleep changes, including their biological substrates, during abstinence is needed to promote effectiveness of abstinence attempts and reduce relapse. Specifically, future work may aim to identify the extent to which sleep changes reinforce cannabis use across various stages of the cannabis addiction lifecycle and the extent that withdrawal-induced sleep disruptions serve as a risk factor for resumption of use during abstinence attempts. Our findings also suggest that future interventions may benefit from incorporating sleep hygiene techniques to improve difficulties with sleep initiation that may emerge in the early days following discontinuation of cannabis use. Use of objective sleep measures, longer follow-up intervals, and high-quality randomized abstinence trials in adolescents with frequent cannabis use are necessary to tailor more effective prevention and cessation efforts for this population.

Supplementary Material

1

Highlights.

  • Cannabis withdrawal is related to self-reported sleep disruption in adolescents

  • Sleep disruption peaked during the first week of abstinence and then decreased

  • Sleep disruption was circumscribed to increases in sleep latency

  • After four weeks of abstinence, sleep seemed comparable to pre-abstinence baseline

  • Interventions could target withdrawal-associated changes in sleep latency

Role of Funding Source:

This work was supported by NIH grant K23DA042946 (to RMS) and K01DA057374 (to JK). The NIH had no role in the study design, collection, analysis, or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of Interest:

None.

Conflict of Interest:

No conflict declared.

References

  1. Adams P, & Barratt E (1975). Effect of chronic marijuana administration of stages of primate sleep-wakefulness. Biological psychiatry, 10(3), 315–322. [PubMed] [Google Scholar]
  2. Adamson SJ, Kay-Lambkin FJ, Baker AL, Lewin TJ, Thornton L, Kelly BJ, & Sellman JD (2010). An improved brief measure of cannabis misuse: the Cannabis Use Disorders Identification Test-Revised (CUDIT-R). Drug and alcohol dependence, 110(1-2), 137–143. [DOI] [PubMed] [Google Scholar]
  3. Allsop DJ, Norberg MM, Copeland J, Fu S, & Budney AJ (2011). The Cannabis Withdrawal Scale development: patterns and predictors of cannabis withdrawal and distress. Drug and alcohol dependence, 119(1-2), 123–129. [DOI] [PubMed] [Google Scholar]
  4. Angarita GA, Emadi N, Hodges S, & Morgan PT (2016). Sleep abnormalities associated with alcohol, cannabis, cocaine, and opiate use: a comprehensive review. Addiction science & clinical practice, 11(1), 1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Arnedt JT, Conroy DA, Stewart H, Yeagley E, Bowyer G, Bohnert KM, & Ilgen MA (2023). Cognitive behavioral therapy for insomnia to reduce cannabis use: Results from a pilot randomized controlled trial. Drug and alcohol dependence, 246, 109835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Babson KA, & Bonn-Miller MO (2014). Sleep disturbances: implications for cannabis use, cannabis use cessation, and cannabis use treatment. Current addiction reports, 1, 109–114. [Google Scholar]
  7. Barratt E, & Adams P (1973). Chronic marijuana usage and sleep wakefulness cycles in cats. Biological psychiatry, 6(3), 207–214. [PubMed] [Google Scholar]
  8. Barratt ES, Beaver W, & White R (1974). The effects of marijuana on human sleep patterns. Biological psychiatry. [PubMed] [Google Scholar]
  9. Bates D, Mächler M, Bolker B, & Walker S (2014). Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:1406.5823. [Google Scholar]
  10. Bolla KI, Lesage SR, Gamaldo CE, Neubauer DN, Funderburk FR, Cadet JL, David PM, Verdejo-Garcia A, & Benbrook AR (2008). Sleep disturbance in heavy marijuana users. Sleep, 31(6), 901–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buysse DJ, Reynolds CF III, Monk TH, Berman SR, & Kupfer DJ (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry research, 28(2), 193–213. [DOI] [PubMed] [Google Scholar]
  12. Cohen-Zion M, Drummond SP, Padula CB, Winward J, Kanady J, Medina KL, & Tapert SF (2009). Sleep architecture in adolescent marijuana and alcohol users during acute and extended abstinence. Addictive behaviors, 34(11), 976–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. D’Souza DC, Cortes-Briones JA, Ranganathan M, Thurnauer H, Creatura G, Surti T, Planeta B, Neumeister A, Pittman B, & Normandin MD (2016). Rapid changes in cannabinoid 1 receptor availability in cannabis-dependent male subjects after abstinence from cannabis. Biological psychiatry: cognitive neuroscience and neuroimaging, 1(1), 60–67. [DOI] [PubMed] [Google Scholar]
  14. Degenhardt L, Coffey C, Romaniuk H, Swift W, Carlin JB, Hall WD, & Patton GC (2013). The persistence of the association between adolescent cannabis use and common mental disorders into young adulthood. Addiction, 108(1), 124–133. [DOI] [PubMed] [Google Scholar]
  15. Di Forti M, Quattrone D, Freeman TP, Tripoli G, Gayer-Anderson C, Quigley H, Rodriguez V, Jongsma HE, Ferraro L, & La Cascia C (2019). The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study. The Lancet Psychiatry, 6(5), 427–436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dijk D-J (2010). Slow-wave sleep deficiency and enhancement: implications for insomnia and its management. The World Journal of Biological Psychiatry, 11( sup1), 22–28. [DOI] [PubMed] [Google Scholar]
  17. ElSohly MA, Mehmedic Z, Foster S, Gon C, Chandra S, & Church JC (2016). Changes in cannabis potency over the last 2 decades (1995–2014): analysis of current data in the United States. Biological psychiatry, 79(7), 613–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Feinberg I, Jones R, Walker J, Cavness C, & Floyd T (1976). Effeets of marijuana extract and tetrahydrocannabinol on electroencephalographic sleep patterns. Clinical Pharmacology & Therapeutics, 19(6), 782–794. [DOI] [PubMed] [Google Scholar]
  19. Freemon FR (1982). The effect of chronically administered delta-9-tetrahydrocannabinol upon the polygraphically monitored sleep of normal volunteers. Drug and alcohol dependence, 10(4), 345–353. [DOI] [PubMed] [Google Scholar]
  20. Gates P, Albertella L, & Copeland J (2016). Cannabis withdrawal and sleep: a systematic review of human studies. Substance abuse, 37(1), 255–269. [DOI] [PubMed] [Google Scholar]
  21. Goodhines PA, Wedel AV, Dobani F, Zaso MJ, Gellis LA, & Park A (2022). Cannabis use for sleep aid among high school students: Concurrent and prospective associations with substance use and sleep problems. Addictive behaviors, 134, 107427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gorelick DA, Goodwin RS, Schwilke E, Schroeder JR, Schwope DM, Kelly DL, Ortemann-Renon C, Bonnet D, & Huestis MA (2013). Around-the-clock oral THC effects on sleep in male chronic daily cannabis smokers. The American journal on addictions, 22(5), 510–514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gruber SA, Dahlgren MK, Sagar KA, Gönenç A, & Lukas SE (2014). Worth the wait: effects of age of onset of marijuana use on white matter and impulsivity. Psychopharmacology, 231, 1455–1465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Halikas JA, Weller RA, Morse CL, & Hoffmann RG (1985). A longitudinal study of marijuana effects. International Journal of the Addictions, 20(5), 701–711. [DOI] [PubMed] [Google Scholar]
  25. Højsgaard S, Halekoh U, & Yan J (2006). The R package geepack for generalized estimating equations. Journal of statistical software, 15, 1–11. [Google Scholar]
  26. Jacobus J, Squeglia LM, Escobar S, McKenna BM, Hernandez MM, Bagot KS, Taylor CT, & Huestis MA (2017). Changes in marijuana use symptoms and emotional functioning over 28-days of monitored abstinence in adolescent marijuana users. Psychopharmacology, 234, 3431–3442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kansagra S. (2020). Sleep disorders in adolescents. Pediatrics, 145(Supplement_2), S204–S209. [DOI] [PubMed] [Google Scholar]
  28. Karacan I, Fernández-Salas A, Coggins W, Carter W, Williams R, Thornby J, Salis P, Okawa M, & Villaume J (1976). Sleep electroencephalographic-electrooculographic characteristics of chronic marijuana users: Part I. Annals of the New York Academy of Sciences, 282(1), 348–374. [DOI] [PubMed] [Google Scholar]
  29. Kaul M, Zee PC, & Sahni AS (2021). Effects of cannabinoids on sleep and their therapeutic potential for sleep disorders. Neurotherapeutics, 18, 217–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kesner AJ, & Lovinger DM (2020). Cannabinoids, endocannabinoids and sleep. Frontiers in molecular neuroscience, 13, 125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kuznetsova A, Brockhoff PB, & Christensen RH (2017). lmerTest package: tests in linear mixed effects models. Journal of statistical software, 82, 1–26. [Google Scholar]
  32. Lisdahl KM, Wright NE, Medina-Kirchner C, Maple KE, & Shollenbarger S (2014). Considering cannabis: the effects of regular cannabis use on neurocognition in adolescents and young adults. Current addiction reports, 1, 144–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lorenzetti V, Hoch E, & Hall W (2020). Adolescent cannabis use, cognition, brain health and educational outcomes: a review of the evidence. European Neuropsychopharmacology, 36, 169–180. [DOI] [PubMed] [Google Scholar]
  34. Lynskey M, & Hall W (2000). The effects of adolescent cannabis use on educational attainment: a review. Addiction, 95(11), 1621–1630. [DOI] [PubMed] [Google Scholar]
  35. Miech R, Johnston L, Patrick M, O’Malley P, Bachman J, & Schulenberg J (2023). Monitoring the future national survey results on drug use, 1975–2022: secondary school students. Ann Arbor: Institute for Social Research, the University of Michigan. [Google Scholar]
  36. Milin R, Manion I, Dare G, & Walker S (2008). Prospective assessment of cannabis withdrawal in adolescents with cannabis dependence: a pilot study. Journal of the American Academy of Child & Adolescent Psychiatry, 47(2), 174–179. [DOI] [PubMed] [Google Scholar]
  37. Pacek LR, Herrmann ES, Smith MT, & Vandrey R (2017). Sleep continuity, architecture and quality among treatment-seeking cannabis users: An in-home, unattended polysomnographic study. Experimental and clinical psychopharmacology, 25(4), 295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Philip P, Dupuy L, Sagaspe P, de Sevin E, Auriacombe M, Taillard J, Micoulaud-Franchi J-A, & Morin CM (2022). Efficacy of a smartphone-based virtual companion to treat insomniac complaints in the general population: sleep diary monitoring versus an internet autonomous intervention. Journal of Clinical Medicine, 11(15), 4387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Polkosnik GL, Sorkhou M, & George TP (2021). Effects of cannabis use on psychotic and mood symptoms: A systematic review. Canadian Journal of Addiction, 12(3), 10–21. [Google Scholar]
  40. Pranikoff K, Karacan I, Larson E, Williams R, Thornby J, & Hursch C (1973). Effects of marijuana smoking on the sleep EEG. Preliminary studies. JFMA, the Journal of the Florida Medical Association. Florida Medical Association, 60(3), 28–31. [PubMed] [Google Scholar]
  41. Quattrone D, Ferraro L, Tripoli G, La Cascia C, Quigley H, Quattrone A, Jongsma HE, Del Peschio S, Gatto G, & Gayer-Anderson C (2021). Daily use of high-potency cannabis is associated with more positive symptoms in first-episode psychosis patients: the EU-GEI case–control study. Psychological medicine, 51(8), 1329–1337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Raniti MB, Waloszek JM, Schwartz O, Allen NB, & Trinder J (2018). Factor structure and psychometric properties of the Pittsburgh Sleep Quality Index in community-based adolescents. Sleep, 41(6), zsy066. [DOI] [PubMed] [Google Scholar]
  43. Schuster RM, Hanly A, Gilman J, Budney A, Vandrey R, & Evins AE (2016). A contingency management method for 30-days abstinence in non-treatment seeking young adult cannabis users. Drug and alcohol dependence, 167, 199–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Schuster RM, Potter K, Vandrey R, Hareli M, Gilman J, Schoenfeld D, & Evins AE (2020). Urinary 11-nor-9-carboxy-tetrahydrocannabinol elimination in adolescent and young adult cannabis users during one month of sustained and biochemically-verified abstinence. Journal of Psychopharmacology, 34(2), 197–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Schweinsburg AD, Nagel BJ, Schweinsburg BC, Park A, Theilmann RJ, & Tapert SF (2008). Abstinent adolescent marijuana users show altered fMRI response during spatial working memory. Psychiatry Research: Neuroimaging, 163(1), 40–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Shrier LA, Burke PJ, Kells M, Scherer EA, Sarda V, Jonestrask C, Xuan Z, & Harris SK (2018). Pilot randomized trial of MOMENT, a motivational counseling-plus-ecological momentary intervention to reduce marijuana use in youth. Mhealth, 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Stockman J (2009). Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review Moore THM, Zammit S, Lingford-Hughes A, et al. (Univ of Bristol, England; et al) Lancet 370: 319–328, 2007. Year Book of Pediatrics, 2009, 570-574. [DOI] [PubMed] [Google Scholar]
  48. Sullivan RM, Wallace AL, Stinson EA, Montoto KV, Kaiver CM, Wade NE, & Lisdahl KM (2022). Assessment of Withdrawal, Mood, and Sleep Inventories After Monitored 3-Week Abstinence in Cannabis-Using Adolescents and Young Adults. Cannabis and Cannabinoid Research, 7(5), 690–699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Suraev A, Mills L, Abelev SV, Arkell TR, Lintzeris N, & McGregor IS (2023). Medical Cannabis Use Patterns for Sleep Disorders in Australia: Results of the Cross-Sectional CAMS-20 Survey. Nature and Science of Sleep, 245–255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Taylor DJ, & Pruiksma KE (2014). Cognitive and behavioural therapy for insomnia (CBT-I) in psychiatric populations: a systematic review. International review of psychiatry, 26(2), 205–213. [DOI] [PubMed] [Google Scholar]
  51. Vandrey R, Budney AJ, Kamon JL, & Stanger C (2005). Cannabis withdrawal in adolescent treatment seekers. Drug and alcohol dependence, 78(2), 205–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Vandrey R, Smith MT, McCann UD, Budney AJ, & Curran EM (2011). Sleep disturbance and the effects of extended-release zolpidem during cannabis withdrawal. Drug and alcohol dependence, 117(1), 38–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wallace ML, Kissel N, Hall MH, Germain A, Matthews KA, Troxel WM, Franzen PL, Buysse DJ, Reynolds C III, & Roecklein KA (2022). Age trends in actigraphy and self-report sleep across the life span: Findings from the Pittsburgh lifespan sleep databank. Psychosomatic Medicine, 84(4), 410–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Watson D, & Clark LA (1991). Mood and anxiety symptom questionnaire. Journal of Behavior Therapy and Experimental Psychiatry. [Google Scholar]
  55. Watson D, Weber K, Assenheimer JS, Clark LA, Strauss ME, & McCormick RA (1995). Testing a tripartite model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. Journal of abnormal psychology, 104(1), 3. [DOI] [PubMed] [Google Scholar]
  56. Yan J. (2002). Geepack: yet another package for generalized estimating equations. R-news, 2(3), 12–14. [Google Scholar]
  57. Yan J, & Fine J (2004). Estimating equations for association structures. Statistics in medicine, 23(6), 859–874. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1939802-supplement-1.docx (18.2KB, docx)

RESOURCES