Treatments for Cannabis Use Disorder Across the Lifespan: A Systematic Review

1. Introduction

Cannabis use disorder (CUD) is characterized by problematic patterns of cannabis use leading to clinically significant impairment or distress [1]. The prevalence of CUD has risen in parallel with increasing potency of cannabis products and growing legalization over the past two decades, now affecting nearly 1.5% of the US population [2,3].

Rates of cannabis use and associated disorders demonstrate considerable variability across different age groups [4,5]. For instance, adolescents and young adults have the highest rates of cannabis use, with over 35% of 12th graders reporting use within the past year [4,5]. Consequently, younger demographic groups are more vulnerable to developing CUD, which can have detrimental effects on neurological development, educational attainment, and psychosocial functioning [6,7]. Among adults, cannabis use peaks during the ages of 18-25, gradually declining thereafter [8]. However, older adults remain susceptible to problematic cannabis use that interferes with occupational, medical, or social domains.

Treatments for CUD have demonstrated relative efficacy, particularly in alleviating withdrawal symptoms, although the effects of existing treatments on relapse prevention and abstinence remain suboptimal. The main modalities of care include pharmacological and psychosocial interventions. Pharmacotherapies such as gabapentin, N-acetylcysteine, and cannabinoid receptor 1 (CB1) agonists have shown promise, while psychosocial approaches, including motivational interviewing (MI), cognitive behavioral therapy (CBT) and contingency management (CM) have demonstrated clinical benefit [9]. Unfortunately, there has been little examination of relative efficacy of these therapeutic interventions across different age groups, especially for adolescents, young, middle-age, and older adults.

Given age-specific variability in the patterns and contexts underlying disordered cannabis use and its consequences, interventions tailored to specific developmental stages may confer superior treatment outcomes. Adolescents and young adults frequently initiate cannabis use recreationally and socially, often lacking insights into problematic behaviors. Thus, psychoeducation and motivational enhancement strategies could potentiate other therapies, for instance. Conversely, older adults more commonly use cannabis under the premise to “self-medicate” underlying psychiatric or medical conditions, suggesting that pharmacological interventions may play a more robust role in treatment outcomes, thus combined strategies targeting both maladaptive cognitions and behaviors besides psychiatric and non-psychiatric comorbidities may be more effective [10].

The comparative effectiveness of current CUD interventions across adolescents, young adults and older adults remains unknown. Our systematic review aimed to examine the literature with hopes to bridge this gap in the existing literature by examining the efficacy of pharmacological and psychosocial interventions for cannabis use disorder across age groups. If successful, findings from our study will help to establish age-specific evidence-based strategies for the treatment of CUD, besides future directions in clinical research to improve the assessment and management of this growing public health concern.

2. Methods

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [11]. The review protocol was registered with PROSPERO (registration number: CRD42024503653).

2.1. Search Strategy

An exhaustive literature search was undertaken to identify studies evaluating pharmacological and non-pharmacological interventions for CUD across adolescent, young adult and older adult populations. We searched the following electronic databases from January 1, 2013, to December 31, 2023: PubMed, Scopus, PsycINFO and ClinicalTrials.gov. The search strategies were developed in collaboration with an experienced medical research librarian with expertise in systematic review searching. The search concepts combined terms related to:

(1) CUD and associated terminology (e.g., “cannabis/marijuana abuse,” “cannabis/marijuana dependence, “cannabis use disorder”) AND (2) Pharmacological interventions, e.g., nabilone, bupropion, gabapentin, cannabidiol, psychedelics “OR” (3) Non-pharmacological (psychosocial and behavioral) interventions for substance use disorders (SUD).

The search strategies were customized for each database using applicable controlled vocabularies and search syntax. The full electronic search strategies utilized for all databases are presented in Appendix A. EndNote X20 reference management software was utilized to collate records retrieved from the literature search.

Eligibility Criteria: Randomized controlled trials (RCTs) that met the following inclusion criteria: (1) Participants diagnosed with CUD, heavy cannabis users, treatment-seeking individuals, or subjects screened as “problematic” cannabis users. (2) RCT assessed any pharmacological (e.g., cannabinoids, gabapentin) or non-pharmacological (e.g., CBT, MI) interventions for treating CUD or reducing problematic use. (3) The study included a comparison group that received placebo, no intervention, standard treatment, or an active comparator. (4) Efficacy outcomes were reported such as reduced cannabis use, abstinence rates, withdrawal symptoms, treatment retention, or cravings.

Exclusion criteria: studies not published in English or not meeting the study design or outcomes of interest as per eligibility criteria.

The inclusion of non-diagnosed, heavy, or problematic cannabis users aimed to enhance the representativeness of our study population. Recognizing that some studies might not explicitly report participants as having CUD despite meeting diagnostic criteria, our approach sought to capture this variability. Documenting quantity/frequency thresholds provided a more consistent measure of severity, while the integration of treatment-seeking behavior and validated screening instruments ensured the identification of clinically significant signs of disordered cannabis use and measurable outcomes for comparison.

The key efficacy outcome measures evaluated across included studies were abstinence, reduction in cannabis use frequency or quantity, withdrawal symptoms, cravings, and treatment retention.

2.2. Study Selection

All records retrieved were imported into Covidence systematic review management software (Veritas Health Innovation, Melbourne, Australia) for screening. After deduplication, two reviewers independently screened titles and abstracts against the eligibility criteria. Any record deemed potentially eligible by either reviewer was retrieved in full-text and re-assessed by both reviewers. Disagreements regarding inclusion/exclusion were resolved by consensus. PRISMA standards guided reporting of the study selection process (Figure 1). A standardized data extraction form was created and pilot-tested. Two reviewers extracted key data (e.g., study design, sample sizes, demographics, diagnostic criteria, interventions, and outcomes of interest). Any discrepancies in extracted data were discussed until consensus was reached.

2.3. Quality Assessment

Two reviewers independently evaluated risk for included RCTs using the Cochrane Risk tool (RoB 2). This tool examines potential issues in five domains: Randomization process, Deviations from intended interventions, Missing outcome data, Measurement of the outcome, and Selection of the reported result. Each domain was judged as low, high, or some concern. Disagreements were resolved through consensus.[12] RCT with low risk across all domains were considered low risk overall, those with some concerns in at least one domain but no high risk were classified as having some concerns overall. The RCT were deemed high risk if they exhibited high risk in at least one key domain or if there were some concerns in multiple domains in a way that substantially lowered confidence in the reported results. The detailed table assessment is available in Supplementary Table S1.

2.4. Data Synthesis

We stratified participants into four age groups: Adolescents (up to 17 years), Young Adults (18-25 years), Transitioning Adults (between young and older adults), and Older Adults (26-65 years). Studies explicitly targeting adolescents or specific adult age ranges were directly classified based on the delineated target population. In instances where inclusion criteria spanned both young and older adulthood, we analyzed the central tendency (mean or median) and dispersion (standard deviation) of reported ages in the population. Studies were categorized as Young Adults or Older Adults if the mean plus/minus one standard deviation fell entirely within the age ranges of 18-25 or 25-65, respectively. Studies where the standard deviation crossed between young adult and older adult ranges were classified as Transitioning Adults. This systematic approach enabled reliable classification of studies across age categories, allowing for targeted comparisons of interventions. Meta-analysis was not conducted due to substantial heterogeneity across studies in interventions, comparators, and outcome measures.

3. Results

3.1. Pharmacological and Neuromodulatory Interventions

To better understand the rationale behind pharmacological and psychosocial interventions covered in this study, a brief review of the current knowledge on the pharmacological and psychoactive effects of cannabis seemed appropriate. The major psychoactive component of Cannabis sativa, Δ-9-tetrahydrocannabinol (THC), acts as a partial agonist at cannabinoid (CB) receptors 1 and 2 [13,14]. Cannabidiol, another derivative of the plant, does not seem to have significant psychoactive properties, although it has shown to have in vitro neuroprotective, anti-inflammatory, and antioxidant properties [15,16,17,18,19]. Additionally, cannabidiol seems to reduce the psychoactive effects of THC by blocking CB1 and CB2 receptors [13,14]. CB1 receptors have been identified as a presynaptic receptors in the central nervous system (CNS), largely in the striatum, amygdalae, hippocampus, cerebellum, and prefrontal cortex, besides peripheral tissues (e.g., gastrointestinal, musculoskeletal, and immune systems) [13,14]. CB2 receptors are found in peripheral tissues, mainly in immune cells [13,14]. The endocannabinoid system seems to be involved in several physiologic functions in the CNS and peripherally, including neuroplasticity and synaptogenesis, cognition, locomotor activity, immune modulation, motivation/hedonic tone, appetite, analgesia, and autonomic function [13,14].

The psychoactive effects of cannabis include relaxation and mild euphoria, although, in some cases, the effects of marijuana can be perceived as unpleasant and associated with anxiety and panic attacks, dysphoria, euphoria, delusions, and hallucinations [20,21,22]. Chronic use of marijuana has been associated with amotivational syndrome, memory impairment, and cognitive decline [22]. Exposure to cannabinoids during brain development seems to affect neuronal circuitry and connectivity, leading to cognitive and behavioral problems, irreversible to some extent [23,24]. Medical complications, including respiratory problems (e.g., chronic bronchitis, airway reactivity and inflammation, increased risk of oropharyngeal and lung cancers by smoking cannabis) are also concerning, some of which worse in geriatric populations [23,25,26].

Noteworthy, cannabis withdrawal symptoms have only recently been recognized by the Diagnostic and Statistical Manual of Mental Disorders (DSM) -5th edition, characterized by irritability, restlessness, anxiety, sleep disruption, anorexia, dysphoria, diaphoresis, abdominal pain, headaches, and body temperature dysregulation [27]. Cannabis withdrawal symptoms can be clinically significant, associated with functional impairment and distress; untreated, these may lead to return to use of cannabis in an attempt to alleviate the reward deficit and stress surfeit associated with this state [27].

Similarly to treating other substance use disorders (SUD), such as nicotine and opioid use disorders with agonist treatments, it has been posited that cannabinoid agonist treatment would not only have the potential to alleviate withdrawal symptoms, but also attenuate the effects of drug use and prevent relapse [28]. To date, numerous preclinical and clinical studies have examined the effects of dronabinol, nabiximols, and cannabidiol for the treatment of CUD [28]. Although the use of cannabinoid agonists for withdrawal symptoms has shown promise, the benefits of these drugs in reducing cannabis use or sustaining abstinence remain uncertain. Small sample sizes, short duration of treatment, heterogeneity among interventions and study outcomes, besides other methodological challenges may represent some of the limitations of current negative findings from RTC to date.

In a randomized double blind clinical trial, Trigo et.al. found that the combination of daily nabiximols (up to 113.4 mg THC/ 105 mg CBD) with motivational enhancement therapy/cognitive behavioral therapy (MET/CBT) was well tolerated and may help reduction of cannabis use [29]. Although cannabis withdrawals and cravings progressively decreased along the 12-week treatment in both groups, there were no significant statistical differences between nabiximols and placebo for the primary outcomes – tolerability and abstinence. In a similar study, Mills et.al. found that counselling combined with nabiximols may provide better outcomes for CUD, where greater attendance to psychosocial interventions was associated with greater odds of abstinence and drug use reduction [30]. Greater pain, disordered sleep, longer exposure to cannabis, and heavier drug use were associated with lower odds of abstinence and engagement in counseling [30]. Interestingly, in this study, higher doses of treatment with nabiximols or placebo were associated with lower odds of reducing frequency of use by 50% and more [30]. Younger age and male sex were associated with lower engagement and poorer outcomes [30].

A double-blind RCT comparing nabiximols (86.4 mg THC and 80 mg CBD) with placebo in a 12-week, multisite trial, found that the placebo group reported significantly more days using cannabis (mean [SD], 53.1 [33.0] days) than the nabiximols group (mean [SD], 35.0 [32.4] days; estimated difference, 18.6 days; 95% CI, 3.5-33.7 days; P = .02), and both groups showed comparable improvements in health status, with no substantial changes in other substance use, suggesting that nabiximols in combination with psychosocial interventions is a safe approach for reducing cannabis use among individuals with cannabis dependence who are seeking treatment [31]. The same research group – The Agonist Replacement for Cannabis Dependence Study group – Arc-D – then, examined the cannabis use outcomes 12 weeks after the initial 12-week intervention. Their findings suggest that the benefits of treatment incorporating nabiximols with psychosocial interventions in reducing cannabis use may persist for up to 3 months after the cessation of treatment [32]. Based on these outcomes, the authors suggested that a short- to medium-term period of treatment with cannabinoid agonists may suffice, particularly if in combination with psychosocial interventions, to sustain longer periods of abstinence, and longer duration of treatment with cannabinoid agonists may be spared for more severe forms of CUD [32].

Allsop et.al. conducted a 6-day placebo-controlled RCT assessing the effects of nabiximols (up to 86.4 mg THC/80 mg CBD) or placebo in combination with standardized psychosocial interventions on severity of cannabis withdrawals and cravings, retention in treatment, and adverse events [33]. In this study, nabiximols attenuated cannabis withdrawal symptoms and improved patient retention in treatment; however, placebo was as effective as nabiximols in promoting long-term reductions in cannabis use following medication cessation [33].

In a double-blind RCT, Levin et.al. examined the effects of dronabinol and lofexidine or placebo on cannabis use and abstinence [34]. in this study, dronabinol augmentation with lofexidine has not shown promise in alleviating withdrawal symptoms and achieving abstinence in CUD [34]. Levin et.al. exploring reductions in self-reported days of cannabis use from three randomized placebo-controlled trials comparing the effects of quetiapine, dronabinol, and lofexidine + dronabinol found that individuals receiving active medications were more likely to move from heavy to moderate use [35]. Here, the authors questioned the clinical significance of moving from heavy to moderate use of cannabis and posited that some medications, such as quetiapine, may be more effective than previously thought if non-abstinent outcomes prove to promote meaningful clinical impact and harm reduction [35].

In the first RCT of CBD for CUD, CBD treatments at 400 mg and 800 mg were safe and more efficacious than placebo at reducing cannabis use by 0.48 days per week (0.15 to 0.82) and 0.27 days per week (-0.09 to 0.64), respectively [36].

Several placebo-controlled trials testing a wide range of pharmacological agents have failed to demonstrate meaningful clinical benefits of these drugs on cannabis outcomes (e.g., withdrawals, cravings, abstinence, use frequency/severity) [28]. Noteworthy, side effects from these off-label treatments are not insignificant, though, often limiting compliance and/or resulting in return to use of cannabis [28].

As aforementioned, the use of quetiapine for the treatment of CUD was associated with an increased likelihood of transitioning from heavy to moderate use, but not light use, although its clinical significance is still unclear [37]. In a double-blind placebo-controlled study, escitalopram treatment did not provide an additional benefit either for achieving abstinence or for the treatment of the cannabis withdrawal syndrome in cannabis dependent individuals [38]. Buspirone treatment has shown to be no more efficacious than placebo in reducing cannabis use [39]. In this study, gender differences were noted, with women in the buspirone treatment group displaying worse cannabis use outcomes [39]. The same research group, in examining the efficacy of vilazodone for treatment of cannabis dependence, found that Vilazodone was not more efficacious than placebo in reducing cannabis use [40]. Important gender differences were also noted here, with worse cannabis use outcomes observed in women [40]. The efficacy of lithium over placebo in the management of cannabis withdrawal was not demonstrated in a placebo-controlled RCT in an inpatient setting [41]. Additionally, in a small trial in which cannabis dependent adults were randomized to receive MET plus oxytocin (n =8) or placebo (n =8), participants receiving oxytocin showed reductions in amount of cannabis used daily and number of sessions per day, effects not observed in the placebo group [42].

In a 12-week double-blind RCT examining the effects of contingency management (CM) and medical management with n-acetylcysteine (NAC) or placebo on cannabis outcomes, there was not statistically significant evidence that the NAC and placebo groups differed in cannabis abstinence (OR=1.00, 95% CI 0.63-1.59, p=0.984) in treatment-seeking adults with CUD [43]. Secondary analysis showed that depressive symptoms did not differ between NAC and placebo groups during treatment, although higher baseline depression was associated with decreased abstinence throughout treatment, suggesting that depressive symptoms may be a risk factor for poorer CUD treatment outcomes. Results from this study did not support the use of NAC to concurrently treat co-occurring depressive symptoms and CUD in adults [44].

Meisel et al. examined the effects MET-CBT plus topiramate (N = 39) or placebo (N = 26) on youths’ time with cannabis-using and non-using friends, cannabis use, and cravings. Their findings suggest that MET-CBT and adjunctive topiramate pharmacotherapy interrupted youth selection processes, and peer affiliations could be one mechanism by which treatments may work [45].

In a pilot study examining the feasibility and effectiveness of varenicline for treating co-occurring cannabis and tobacco use found varenicline to be well-tolerated and may reduce cannabis craving, cannabis use, and tobacco use in this population [46]. In a proof-of-concept 6-week randomized, placebo-controlled pilot trial examining the effects of either varenicline (n = 35) or placebo (n = 37), added to a brief MET found significant decreases in self-reported cannabis withdrawal, percentage of days used, and use per day in both groups during treatment compared to baseline. In this study, the authors hypothesized feasibility of conducting clinical trials of varenicline as a candidate pharmacotherapy for CUD [47].

Interestingly, a recent RCT found that, relative to placebo, naltrexone maintenance treatment significantly reduced both active cannabis self-administration and subjective reinforcing effects in non-treatment-seeking daily cannabis smokers [48]. lastly, In a double-blind, sham-controlled, crossover design, a single session of repetitive transcranial magnetic stimulation (rTMS) applied to the dorsolateral pre-frontal cortex (DLPFC) although well tolerated, failed to reduce cue-elicited craving in heavy cannabis users [49].

3.2. Psychosocial and Behavioral Interventions

Advances in psychosocial and behavioral interventions for SUDs have been observed over the past decades [50,51,52]. These non-pharmacological interventions can be used alone or in combination with pharmacotherapy. Cognitive behavioral therapy (CBT), contingency management (CM), motivational interviewing (MI)/motivational enhancement treatment (MET), mindfulness-based relapse prevention (MBRP), community reinforcement approach (CRA), and matrix model therapy (MMT) are examples of well-established treatments for SUDs [50,51,52].

CBT emphasizes the importance of correcting cognitive distortions to improve behaviors and emotions related to substance use [51]. It involves a combination of strategies via operant learning, cognitive and motivational strategies, and skills-building interventions. [51] MBRP also appears to be effective for treatment of SUDs [52,53]. By increasing awareness and acceptance through nonjudgmental approaches, individuals tend to experience lower levels of cravings, potentially enhancing clinical outcomes [52,53]. Through enhancing intrinsic motivation to change by exploring and resolving ambivalence about substance use, MI has been proven an effective approach to treat SUD [54]. CM and Community Reinforcement Approach (CRA) are outpatient programs of proven efficacy for treatment of SUDs, associated with longer treatment adherence, longer periods of abstinence, and greater improvements in functioning [55,56,57,58]. Based on reward strategies to reinforce abstinence and positive behaviors, users are usually given incentives to remain engaged in treatment and abstain from drug use. Additionally, the MMT is a structured and usually time-limited approach combining individual and group therapy, family psychoeducation and social support to improve adherence and facilitate recovery [59,60,61]. Facilitating adaptive social networks seem to also play an important role in social functioning, leading to overall better SUD outcomes [62]. Approach bias modification, a novel psychosocial intervention that intends to attenuate approach bias and the incentive salience of drug has shown promise for relapse prevention in SUD and binge eating behaviors [63].

Multiple studies have explored the effects of these interventions for specific age groups, and the results are displayed in Table 1. Unfortunately, limited access to the above psychosocial treatments due to financial barriers, treatment availability, particularly in rural areas, low motivation for treatment, and limited insights into severity of drug use, besides data suggesting deviation from evidence-based treatments by clinicians, seem to greatly impact clinical outcomes and overall prognosis in SUD clinical course and treatment [64].

Novel approaches, including technology-based interventions, particularly smartphone applications, have gained popularity in recent years and may represent accessible and cost-effective alternative methods of treatment delivery for SUDs [65].

4. Discussion

The use of cannabinoids for recreational and therapeutic purposes has been described for centuries [66]. Cannabis accounts for the third most commonly used substance worldwide, only after alcohol and tobacco [67]. Considering that around 1 in 10 of regular users of cannabis will develop moderate to severe forms of CUD over time, the need for more effective treatments for this condition is pressing [67]. Moreover, the increasing prevalence of cannabis use, CUD, and its complications over the past decade is not negligible, and treatments to date remain insufficient. As of now, there are no FDA-approved medications for this condition, and the off-label use of psychotropic medications has only demonstrated modest to no benefits, particularly for relapse prevention and abstinence [28]. At best, a few psychotropic medications, including cannabinoid agonists, have shown the potential to alleviate cannabis withdrawal symptoms, particularly insomnia, anorexia, and anxiety/restlessness, whereas most placebo-controlled trials for CUD testing a wide range of psychotropic agents have failed to demonstrate benefits for relapse prevention and sustained abstinence [28]. Unfortunately, side effects from these off-label treatments are not insignificant, often limiting compliance and/or resulting in return to use of cannabis. Moreover, the use of these medications for specific age-groups, such as adolescents and older adults, remain unclear, since most studies tend to exclude these specific cohorts from trials. Behavioral and psychosocial interventions remain the main stay treatment for CUD. However, limited access to these evidence-based interventions, engagement, and retention in treatment are often suboptimal, impacting clinical outcomes and overall prognosis.

Notably, while some studies have explored correlates of treatment outcomes, the paucity of data on age-specific treatments for CUD in the existing literature remains. Noteworthy, data suggest that SUDs in older adults remain underestimated and largely untreated, in part because this population has been traditionally accounted for only a small fraction of the problem [68,69,70]. The implications of underdiagnosing and undertreating older adults with SUD, including CUD, are particularly concerning, since these individuals are more vulnerable to adverse outcomes associated with drug use [71,72].

Despite an increasing need for older adult substance use services, facilities with programs designed for older persons remain relatively scarce [73]. In a study of 13,749 responding facilities in the U.S., only 17.7% had specific programs for older adults [73]. Noteworthy, evidence demonstrates that treatments for SUDs are cost-effective and tend to have similar rates of recurrence/relapse compared to other chronic illnesses in the older adult population [74]. Moreover, evidence has also shown that older individuals tend to have greater adherence to treatment and clinical outcomes, including days of use and abstinence rates, regardless of level of care, compared to younger counterparts [75]. Unfortunately, screening, diagnosis, and treatment of SUDs in older individuals remain suboptimal [76,77]. Expanding substance use services, particularly in primary care settings, would likely be considerably impactful, since primary care providers may play an important role in early detection and delivery of brief interventions particularly for this particular age group [78,79].

While psychosocial interventions may have greater specificity based on individual characteristics, such as age and other psychosocial determinants of health, the existing literature does not provide enough evidence to support age-specific pharmacological treatments for CUD. Therefore, the use of off-label agents for the treatment of CUD requires great caution and personalized assessments. Individual factors, such as age and age-related changes in pharmacokinetics and pharmacodynamics, medical and psychiatric comorbidities, drug-drug interactions, and side effect profile may play a critical role in treatment recommendations and off-label use of medications.

The need for greater access to care and more effective, affordable, and age-appropriate treatments for CUD remains. Novel treatments, including technology-based approaches, neuromodulatory interventions, and pharmacological treatments have shown promise, although future research is warranted. Through advances in understanding the pathophysiological underpinnings of CUD and age-specific variabilities, novel pharmacological and psychosocial treatments may be attainable. Lastly, while strategies to promote harm reduction have been implemented since the mid-1990s, particularly with the advent of the opioid epidemic, at this time, it remains unclear whether the change from heavy to moderate use of cannabis, for instance, will result in clinically meaningful outcomes and long-term impact [35]. Future research is needed to elucidate this question.

5. Conclusions

Cannabis use is a growing problem for which pharmacological treatments and psychosocial interventions remain insufficient and often inadequate. As of now, there are no FDA-approved medications for this condition, besides insufficient evidence to support the clinical utility of off-label use of any specific medication or age-specific treatments. Presently, behavioral and psychosocial interventions are the main stay treatment for CUD, and novel treatments, including technology-based approaches, have shown promise. Given considerable variability among age groups, age-specific treatments need to be further explored. As it is still uncertain whether harm reduction strategies will result in clinically meaningful outcomes in CUD, future research is warranted to elucidate this question. Education, prevention, and early interventions are critical, especially considering the public’s perception of low harm and widespread benefits from this drug. Cannabis is not a panacea!