Effectiveness and safety of psychosocial interventions for the treatment of cannabis use disorder: a systematic review and meta-analysis

Monika Halicka, Thomas L Parkhouse, Katie Webster, Francesca Spiga, Lindsey A Hines, Tom P Freeman, Sabina Sanghera, Sarah Dawson, Craig Paterson, Jelena Savović, Julian PT Higgins, Deborah M Caldwell
doi: https://doi.org/10.1101/2024.11.18.24317475

ABSTRACT

Aim To evaluate the effectiveness, safety, and cost-effectiveness of psychosocial interventions for cannabis use disorder (CUD).

Design A systematic review of randomized controlled trials (RCTs; PROSPERO protocol CRD42024553382). We searched databases (MEDLINE/PsycInfo/Cochrane CENTRAL) to 12-Jun-2024. We assessed results using Risk of Bias 2 and conducted meta-analyses where possible.

Setting Inpatient/outpatient/community-based.

Participants Individuals with CUD aged ≥16 years.

Interventions Psychosocial interventions lasting >4 sessions, delivered in real time.

Measurements Primary outcomes were continuous- and point-abstinence, withdrawal intensity, treatment completion and adverse events.

Findings We included 22 RCTs (3,304 participants). At the end of treatment, cognitive-behavioural therapy (CBT) increased point abstinence (odds ratio [OR]=18.27, 95% confidence interval [9.00; 37.07]), and continuous abstinence (OR=2.72 [1.20; 6.19]), but reduced treatment completion (OR=0.53 [0.35; 0.85]) versus inactive/nonspecific comparators. Dialectical behavioural/acceptance and commitment therapy increased point abstinence versus inactive/nonspecific comparators (OR=4.34 [1.74; 10.80]). The effect of CBT plus affect management versus CBT on point abstinence was OR=7.85 [0.38; 163.52]. The effect of CBT plus abstinence-based contingency management versus CBT on point abstinence was OR=3.78 [0.83; 17.25], and on continuous abstinence OR=1.81 [0.61; 5.41]. For CBT plus abstinence-contingency management versus CBT plus attendance-contingency management, the effect on point abstinence was OR=1.61 [0.72; 3.60], and on continuous abstinence OR=2.04 [0.75; 5.58]. The effect of community reinforcement on point abstinence was OR=0.29 [0.04; 1.90] versus CBT, and on continuous abstinence OR=47.36 [16.00; 140.21] versus nonspecific comparator. Interventions other than CBT may not affect treatment completion. No adverse events were reported. No study reported withdrawal intensity. Two economic evaluations reported higher costs for more complex psychosocial interventions and contingency management.

Conclusions Cognitive-behavioural and dialectical behavioural/acceptance and commitment therapies may increase abstinence relative to inactive/nonspecific comparators. The conclusions remain tentative due to low to very low certainty of evidence and small number of studies.

BACKGROUND

Worldwide, cannabis is the most widely used illicit drug. In 2022, the number of people aged 15-64 years using cannabis was estimated as 228 million, representing 4.4% of the global population.1 The risk of developing dependence on cannabis significantly increases with increasing frequency of use.2 The diagnosis of cannabis use disorder (CUD) outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5),3 requires the presence of at least two of eleven criteria. These include hazardous use, social or interpersonal problems related to use, neglect of major roles, withdrawal, tolerance and cravings, amongst other features. DSM-5 CUD amalgamated previous diagnoses of cannabis dependence and cannabis abuse included in DSM-IV.4 Cannabis dependence is also listed in the International Classification of Diseases (ICD).5

The prevalence of CUD has been escalating globally, affecting over 15 million men and over 8 million women in 2019.6 The burden of CUD is the highest among young adults aged 20-24 years.6 The number of people enrolling in treatment for cannabis use has been increasing globally7 and it is also the most frequently cited problem drug among people entering drug treatment.1,8 For instance, across Europe, number of adults seeking treatment increased from 27 per 100,000 in 2010 to 35 per 100,000 in 2019.9

According to the World Health Organization (WHO), adults using cannabis should be offered brief interventions, focused on individualized feedback and advice.10 However, for people diagnosed with CUD or dependence, such brief interventions have limited benefit and WHO recommend they should be referred for specialist cannabis-specific treatment.10 Psychosocial interventions (PSIs) are, currently, the only recommended treatment for people with CUD.7,11,12 This recommendation is supported by evidence from systematic reviews that suggest PSIs are effective for treatment of CUD.13–17 In contrast, evidence for the use of pharmacological treatments for CUD is lacking.18 However, these previous reviews provide limited insight into the specific types of PSIs that are most effective for treating CUD. For example, some reviews aggregate various types of PSIs for comparison against inactive controls in pairwise meta-analysis,13,14 some provide a descriptive summary of results from individual randomized controlled trials (RCTs),17 while others report an overview of findings from published systematic reviews.19,20

Quantitative estimates of intervention effect, safety, and cost-effectiveness are important to inform policy and clinical decision-making. However, to date, reviews have not included safety outcomes or economic evaluations of PSIs for treatment of CUD. The purpose of the present review is to provide an up-to-date and rigorous review of the evidence for clinical effectiveness, safety, and cost-effectiveness of PSIs for the treatment of CUD in adults and young people aged ≥16 years.

METHODS

The review protocol was prospectively registered with PROSPERO (CRD42024553382).21 The review is reported following PRISMA guidelines.22

Eligibility criteria

Study eligibility criteria are presented in Table 1. PSIs were grouped based on shared theoretical underpinning and the therapeutic techniques used. Intervention and comparator categories are summarised in Table 2, with more detail provided in Supporting Information 1. Explanation of outcome operationalization and the hierarchy of preference followed for studies reporting multiple measures and/or follow-up timepoints, are outlined in Supporting Information 2.

View this table:
Table 1. Eligibility criteria
View this table:
Table 2. Intervention and comparator categories

Searches

We searched Ovid MEDLINE-ALL and PsycInfo, and the Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library using relevant subject headings, text-words and search syntax appropriate to each resource (all available years to 12-Jun-2024). Reports of RCTs from Embase and CINAHL were captured via our search of CENTRAL.23 To identify potentially relevant economic evaluations, we ran separate searches in Ovid MEDLINE and Embase (all available years to 30 July 2024). Search strategies are provided in Supporting Information 3.

Study selection

Titles and abstracts were screened independently by at least two reviewers using the Rayyan platform.24 Potentially relevant texts were retrieved in full, and assessed independently by at least two reviewers using the LaserAI platform.25 Discrepancies were resolved by discussion with a third reviewer, or the wider review team.

Data extraction

Using piloted, standardized forms created in LaserAI,25 we extracted details on study design and conduct, eligibility criteria, participant demographics and PROGRESS-Plus characteristics,26 intervention and comparator details, and outcome data. We used LaserAI’s AI-enhanced suggestions to support extraction of study characteristics.25 However, all suggestions were verified and amended, if needed, by a reviewer.

Arm-level numerical data for dichotomous outcomes were extracted as the number of participants with event, number with available outcome data, and number randomized into each arm (the denominator used in the analysis depended on outcome – see synthesis of results). For continuous outcomes, mean (M) with standard deviation (SD) and number of participants analysed in each arm were extracted for end of treatment and follow-up. Data processing steps in preparation for synthesis are outlined in Supporting Information 4.

Study characteristics and numerical data were extracted by a single reviewer and checked in detail by a second reviewer. Discrepancies were resolved through discussion, or with a third reviewer.

Risk of bias assessment

We assessed risk of bias (RoB) using the RoB227 tool at the outcome level for each study, for all primary and secondary effectiveness outcomes reported at the end of treatment, and safety outcomes related to adverse events at any time. RoB2 was assessed initially by two reviewers independently. Once consistency had been achieved, RoB2 was assessed by one reviewer and checked by a second. Two reviewers independently assessed economic outcomes using the Drummond and Jefferson critical appraisal checklist.28

Synthesis of results

Pairwise random-effects meta-analyses were conducted in R software version 4.3.1, using ‘meta’ package version 7.0-0.29–31 Effect estimates were pooled if there were at least two studies contributing data for the same comparison; otherwise, study-level effect estimates are presented. Statistical heterogeneity was assessed using the I2 statistic to quantify inconsistency, with strength of evidence quantified using the p-value from the chi2 test. The between-study variance, τ2, was estimated using the restricted maximum-likelihood (REML) method. We assumed a common τ2 across all comparisons within the same outcome and timepoint (as is done in a network meta-analysis).32 This was because there were too few studies to estimate τ2 reliably within each comparison33 and we had no reason to expect that the between-study variance would differ across comparisons. To estimate a common τ2, we used residual τ2 from a meta-regression with comparison included as a covariate. A fixed continuity correction (0.5) was added to studies with zero events in one arm. We present the results as odds ratios (ORs) for dichotomous outcomes and ratios of means (RoMs) for continuous outcomes,34–36 with 95% confidence intervals (CIs).

Primary meta-analyses were conducted at the end of treatment. Additional timepoints were medium (≤6 months) and long follow-up (>6 months post-treatment). Analyses were based on the number of participants with available outcome data, except for completion of treatment which was based on the number of participants randomized. To minimize multiplicity of analyses (e.g. by reusing the same intervention arms across different comparisons), the comparisons reported in the Results section were prioritised for synthesis, and study-level effect estimates for other potential comparisons are presented in Supporting Information 5.

Sensitivity and subgroup analyses

Sensitivity analyses addressed fixed-effect meta-analyses, and imputing missing outcome data as abstinent or non-abstinent for dichotomous abstinence outcomes. We planned subgroup analyses to explore heterogeneity using the following potential effect modifiers: intensity and duration of cannabis use, mental health co-morbidities, intervention intensity, treatment setting, use of adjunct interventions or booster sessions, and PROGRESS-Plus characteristics.26

Certainty of evidence

We used the GRADE framework37 to assess the certainty of evidence for effectiveness of PSIs on primary and secondary outcomes at the end of treatment. There are no established thresholds representing minimal clinically important differences for these outcomes. In this review, we describe intervention effects as clinically meaningful if they represent a 10% increase or reduction of risk in the intervention group relative to the comparator for dichotomous outcomes. For continuous outcomes effects are assessed as clinically meaningful if an intervention halves or doubles the frequency, quantity, or duration of outcome relative to a comparator. The criteria considered for grading the certainty are outlined in Supporting Information 6.

RESULTS

Included studies

Thirty-two reports of k=22 studies (participants n=3,304) were included (Figure 1). Details of excluded reports are listed in Supporting Information 7.

Figure 1.
Figure 1.

PRISMA flow diagram. Citation searching did not identify any additional records, therefore, identification of studies via other methods is not displayed.

Characteristics of included studies are presented in Table 3. Study sample size ranged from 40 to 450 participants (M=150, SD=103). Most studies were conducted in the United States (k=15) and in outpatient settings (k=15). Participants were mostly male (M=80%, range 56-100%) and of white ethnicity. Mean age ranged from 16 to 48 years (M=28, SD=8). Education ranged from secondary school to degree-level (k=10) or mean 13.5 years of education (k=7), most participants were employed (k=12), and on average 36% were married or co-habiting (k=10). Several studies excluded individuals with excessive commuting distance/transportation difficulties or unstable living situations (k=5), those with serious mental health issues (k=14), or those who required inpatient treatment or had serious medical problems (k=8). A detailed summary of the PROGRESS-Plus26 characteristics relating to equity is presented in Supporting Information 8. Participants met diagnostic criteria for CUD (k=5), cannabis dependence (k=10), abuse (k=1), dependence or abuse (k=4), or a diagnostic screening cut-off (k=2). On average, participants used cannabis on 74% of days (SD=21%, k=15).

View this table:
Table 3. Characteristics of included studies

CBT was the most commonly evaluated PSI (k=15), followed by abstinence-based contingency management (CM-abstinence; k=8), attendance-based CM (CM-attendance; k=4), dialectical behavioural/acceptance and commitment therapies (DBT/ACT; k=3), CBT with affect management (CBT-affect; k=2), community reinforcement (k=2), and multidimensional family therapy (MDFT; k=1). Inactive/nonspecific comparators were used in k=13 studies. Interventions were delivered over 1-6 months (M=2.77, SD=1.09) and consisted of 6-52 sessions (M=13.84, SD=10.81). Most sessions occurred weekly (60%). All interventions were delivered in person, either individually (61%), as a group (19%), or mixed (10%) format (10% were not reported).

Included studies reported point- (k=10), and continuous-abstinence (k=12), duration of continuous abstinence (k=8), completion of treatment (k=16), frequency (k=17) and quantity of cannabis use (k=3), craving (k=1), adverse events (k=2), and cost-effectiveness outcomes (k=2) at any timepoint. No studies reported on intensity of withdrawal, engagement in further treatment, or dropout due to adverse events.

Risk of bias

Across all effectiveness outcomes at the end of treatment and safety outcomes, we judged 70% to be at high RoB, 21% to have some concerns, and only 9% to be at low risk of bias. The main concerns were bias in selection of the reported result (e.g. lack of pre-specified analysis plan), bias due to missing outcome data (e.g. high attrition likely dependant on participant relapse), and bias in measurement of outcome (e.g. self-report by unblinded participants). Supporting Information 9 includes detailed assessments for each RoB2 domain.

Results synthesis

The effectiveness results reported below are for the end of treatment timepoint only (4-24 weeks). Results for medium and long follow-up timepoints are reported in Supporting Information 10. Subgroup analyses were not possible due to an insufficient number of studies, but the relevant characteristics are reported in Table 3. Sensitivity analyses are reported in Supporting Information 11. Summary of findings tables are presented in Supporting Information 12. Interpretation of findings is based on minimally important clinical differences, and takes into account the GRADE assessments of certainty of evidence. Interpretations are not based on statistical significance.38

Point abstinence

Nine studies39–47 were included in the analysis for point abstinence. Seven measured abstinence using urine tests, one with self-report,39 and one used both.46 Two studies defined point abstinence as seven days of abstinence.39,40 Evidence of effectiveness is of very low certainty due to concerns over high RoB and imprecision, and for some comparisons also due to indirectness (Supporting Information 12). The common τ2 was estimated as 0.00 (standard error, SE=0.38). Meta-analyses included a maximum of two studies per comparison (Figure 2). CBT relative to a waitlist comparator (OR=18.27, 95% CI [9.00; 37.07]), and DBT/ACT relative to a nonspecific comparator (OR=4.34 [1.74; 10.80]), may lead to clinically meaningful increases in point abstinence. CBT plus CM-abstinence may improve abstinence compared with CBT (OR=3.78 [0.83; 17.25]) but the CIs are also consistent with a decrease in abstinence (i.e. favouring CBT). There is little to no evidence of an effect of CM-abstinence relative to CM-attendance when both are delivered with CBT (OR=1.61 [0.72; 3.60]). Community reinforcement may be associated with a meaningful decrease in abstinence when compared with CBT (OR=0.29 [0.04; 1.90]), although the CIs are also consistent with an increase in abstinence. The comparison of CBT-affect relative to CBT is based on a single study with zero events in the comparator group and the effect estimate is highly uncertain (OR=7.85 [0.38; 163.52]).

Figure 2.
Figure 2.

Forest plot for random-effects meta-analyses of point abstinence at end of treatment. ACT, acceptance and commitment therapy; BT, behavioural therapy; CBT, cognitive-behavioural therapy; CBT-I, CBT for insomnia; CI, confidence interval; CM-ab/at, contingency management based on abstinence/attendance; ComReinf, community reinforcement; DBT, dialectical behavioural therapy; ICART, integrated cannabis and anxiety reduction treatment; MET, motivation enhancement therapy; NS, nonspecific comparator; OR, odds ratio; Wait, waitlist. Risk of bias (A) arising from the randomization process, (B) due to deviations from intended interventions, (C) due to missing outcome data, (D) in measurement of the outcome, (E) in selection of the reported result, (O) overall; ‘+’, low risk, ‘?’, some concerns, ‘-‘, high risk of bias.

Continuous abstinence

Ten studies measured continuous abstinence (lasting 6-14 weeks) up to the end of treatment.44,45,47–54 Most used self-report measures, with only two using consecutive negative urine tests,44,47 and one verifying self-reports with urine tests.50 Evidence of effectiveness is of low to very low certainty due to concerns over high RoB, imprecision and inconsistency (see Supporting Information 12 for comparison-specific assessments). Meta-analyses included up to four studies per comparison (Figure 3). The common τ2 was estimated as 0.49 (SE=0.42). CBT may increase continuous abstinence relative to inactive/nonspecific comparators (OR=2.72 [1.20; 6.19]). This analysis is characterized by high heterogeneity (I2=82%) that may be explained by comparator type (waitlist or nonspecific). CBT plus CM-abstinence may increase continuous abstinence relative to CBT plus CM-attendance, although CIs are also consistent with a decrease in abstinence (i.e. favouring CBT plus CM-attendance; OR=2.04 [0.75; 5.58]). There is little to no evidence of an effect of CBT plus CM-abstinence relative to CBT (OR=1.81 [0.61; 5.41]). The comparison of community reinforcement versus nonspecific comparator is based on one study54 with unclear definition of continuous abstinence and the effect estimate is highly uncertain (OR=47.36 [16.00; 140.21]).

Figure 3.
Figure 3.

Forest plot for random-effects meta-analyses of continuous abstinence at end of treatment.

BT, behavioural therapy; CBT, cognitive-behavioural therapy; CI, confidence interval; CM-ab/at, contingency management based on abstinence/attendance; ComReinf, community reinforcement; MET, motivation enhancement therapy; NS, nonspecific comparator; OR, odds ratio; RelPrev, relapse prevention; Wait, waitlist.

Risk of bias (A) arising from the randomization process, (B) due to deviations from intended interventions, (C) due to missing outcome data, (D) in measurement of the outcome, (E) in selection of the reported result, (O) overall; ‘+’, low risk, ‘?’, some concerns, ‘-‘, high risk of bias.

For the related outcome of mean duration of continuous abstinence, there is very low certainty evidence of little to no effect for CBT versus nonspecific comparator, CBT plus CM-abstinence versus CBT, and CBT plus CM-abstinence versus CBT plus CM-attendance (RoMs range 1.24-1.40; Supporting Information 10).

Completion of treatment

The number of participants who completed treatment was reported in 16 studies.39,41–43,45–48,50–52,54–58 Meta-analyses included a maximum of five studies (Figure 4). The common τ2 was estimated as 0.00 (SE=0.13). There is low certainty evidence that CBT may be associated with lower completion rates than inactive/nonspecific comparators (OR=0.53 [0.35; 0.82]). We found low certainty evidence for CBT plus CM-abstinence compared with CBT (OR=1.58 [0.85; 2.94]), and for community reinforcement relative to CBT or nonspecific comparator (OR=1.20 [0.49; 2.96]). The certainty of evidence is very low for DBT/ACT versus inactive/nonspecific comparators (OR=1.42 [0.59; 3.43]), CBT-affect versus CBT (OR=1.03 [0.45; 2.32]), and CBT plus CM-abstinence versus CBT plus CM-attendance (OR=1.12 [0.48; 2.62]). We had concerns over indirectness across all comparisons, and RoB and imprecision for most comparisons (Supporting Information 12).

Figure 4.
Figure 4.

Forest plot for random-effects meta-analyses of completion of treatment.

ACT, acceptance and commitment therapy; AMT, affect management therapy; CBT, cognitive-behavioural therapy; CBT-I, CBT for insomnia; CI, confidence interval; CM-ab/at, contingency management based on abstinence/attendance; ComReinf, community reinforcement; DBT, dialectical behavioural therapy; ICART, integrated cannabis and anxiety reduction treatment; MET, motivation enhancement therapy; Mind/Edu, mindfulness psychoeducation; NS, nonspecific comparator; OR, odds ratio; RelPrev, relapse prevention; Wait, waitlist.

Risk of bias (A) arising from the randomization process, (B) due to deviations from intended interventions, (C) due to missing outcome data, (D) in measurement of the outcome, (E) in selection of the reported result, (O) overall; ‘+’, low risk, ‘?’, some concerns, ‘-‘, high risk of bias.

Frequency of cannabis use

Sixteen studies39,41,42,44,45,47–53,55,57–59 reported frequency of use measured over past 7-90 days. Fifteen used self-report, and one used weekly urine tests.55 Meta-analyses included up to six studies per comparison (Figure 5). The common τ2 was estimated as 0.08 (SE=0.05). The evidence is of very low certainty due to high RoB for all comparisons, and due to imprecision, indirectness, and inconsistency among some comparisons (Supporting Information 12). DBT/ACT may have a clinically meaningful effect on reducing frequency of use relative to nonspecific comparators (RoM=0.39, 95% CI [0.25; 0.60]). For other intervention comparisons there is little to no evidence of an effect, i.e., none were estimated to halve or double the frequency of use. This includes CBT versus inactive/nonspecific comparators (RoM=0.63 [0.48; 0.83]), CBT-affect versus CBT (RoM=0.93 [0.56; 1.55]), CBT plus CM-abstinence versus CBT (RoM=0.88 [0.65; 1.19]), CBT plus CM-abstinence versus CBT plus CM-attendance (RoM=0.98 [0.68; 1.40]), and MDFT versus CBT (RoM=0.81 [0.69; 0.95]). The analysis of CBT versus inactive/nonspecific comparators is characterized by high heterogeneity (I2=83%) that may be explained by comparator type.

Figure 5.
Figure 5.

Forest plot for random-effects meta-analyses of frequency of cannabis use at the end treatment. Frequency of use is expressed as proportion of days using for most studies, except for proportion of weeks using in Carroll 2006, and number of uses in NCT02102230.

ACT, acceptance and commitment therapy; AMT, affect management therapy; BT, behavioural therapy; CBT, cognitive-behavioural therapy; CBT-I, CBT for insomnia; CI, confidence interval; CM-ab/at, contingency management based on abstinence/attendance; DBT, dialectical behavioural therapy; MDFT, multidimensional family therapy; MET, motivation enhancement therapy; NS, nonspecific comparator; RelPrev, relapse prevention; ROM, ratio of means; Wait, waitlist. Risk of bias (A) arising from the randomization process, (B) due to deviations from intended interventions, (C) due to missing outcome data, (D) in measurement of the outcome, (E) in selection of the reported result, (O) overall; ‘+’, low risk, ‘?’, some concerns, ‘-‘, high risk of bias.

Quantity of cannabis use

Two studies reported quantity of cannabis use. This evidence is of very low certainty due to concerns over high RoB, imprecision and, for Buckner 201943, indirectness. Babor 202451 measured self-reported number of joints smoked per typical day of use over past 90 days. The results suggest that CBT may have a clinically meaningful effect on reducing quantity of use when compared with a waitlist control (RoM=0.49 [0.35; 0.69]). Buckner 201943 measured self-reported total number of joints smoked over past 30 days. The evidence indicates that CBT-affect may reduce the quantity of use when compared with CBT (RoM=0.49 [0.17; 1.38]), although CIs are also consistent with an increase in the quantity of use.

Craving

A single study measured current cannabis craving, using Marijuana Craving Questionnaire short-form (Davoudi 2021a).41 Results indicate there may be little to no evidence of an effect of DBT on reducing craving relative to a nonspecific comparator (RoM=0.95 [0.86; 1.04]; very low certainty evidence due to RoB, imprecision, and indirectness).

Adverse events

Two studies reported adverse events.39,45 Stanger 200945 compared CBT plus CM-abstinence with CBT plus CM-attendance and NCT02102230 201439 compared CBT with a nonspecific comparator. Certainty of evidence is very low due to concerns over RoB, imprecision, and indirectness. For both studies, we could not estimate intervention effects due to the lack of adverse events in either group.

Cost-effectiveness outcomes

We identified two trial-based economic evaluations,60,61 both carried out from a healthcare perspective. Full details of these studies and accompanying critical appraisal are presented in Supporting Information 13. Goorden 201660 conducted cost-effectiveness analyses of MDFT compared with CBT for adolescents, based on a single Dutch site of a multicentre trial.59 Over a 12-month time horizon, MDFT was associated with higher costs than CBT, better quality of life (EQ-5D-3L), and increased recovery rate (difference in recovery was not statistically significant). The incremental cost-effectiveness ratio (ICER) was €54,308 per quality-adjusted life years (QALYs) gained, and €43,405 per additional recovered patient.

Olmstead 200761 conducted cost-effectiveness analyses of CBT plus CM-abstinence/attendance, CBT alone, CM-abstinence/attendance, and nonspecific comparator (counselling) for young adults referred by the criminal justice system, based on a multi-arm trial55 from the United States. Over the initial eight-week treatment period and an additional eight-month time horizon, interventions did not differ significantly in effectiveness, but the costs were the highest for CBT plus CM, followed by CM, CBT, and nonspecific comparator. ICERs for CBT relative to nonspecific comparator were $102 USD per additional week of continuous abstinence (reduced to $34 USD at follow-up) and $159 USD per additional negative urine sample. For CM relative to CBT, ICER was $1104 USD per additional week of continuous abstinence. For CBT plus CM relative to CBT, ICERs were $1333 USD per additional week of continuous abstinence (reduced to $915 USD at follow-up), and $942 USD per additional negative urine sample.

Effectiveness outcomes at follow-up

Based on a smaller number of studies, there is some evidence that beneficial effects of CBT and DBT/ACT versus inactive/nonspecific comparators for point abstinence, of CBT plus CM-abstinence versus CBT plus CM-attendance for point and continuous abstinence, and of community reinforcement versus other interventions for continuous abstinence, may be maintained up to six months post-treatment. The relative advantages of the latter two may still be present over six months post-treatment (Supporting Information 10).

Sensitivity analyses

Results of the sensitivity analyses using fixed-effect meta-analyses were broadly consistent with random-effects across all effectiveness outcomes. Sensitivity analyses imputing missing observations as abstinent at the end of treatment suggest the effects of some PSIs on point abstinence may be reduced, but remain similar for continuous abstinence. Imputing missing observations as non-abstinent had little impact on the results (Supporting Information 11).

DISCUSSION

This review aimed to evaluate the effectiveness, safety, and cost-effectiveness of PSIs for CUD in people aged ≥16 years. We included 22 RCTs (3304 participants). We judged the certainty of the evidence to be low to very low, due to concerns of high RoB and imprecision of the estimated treatment effects. Various effectiveness outcomes were reported across the studies and only two reported safety outcomes.

CBT-based interventions were the most commonly evaluated PSI, followed by CM and DBT/ACT. At the end of treatment, we found CBT led to a clinically meaningful benefit for point and continuous abstinence but was associated with lower treatment completion. CBT-affect and CBT plus CM-abstinence compared with CBT may have clinically meaningful effects on point abstinence, but there was no evidence that CBT plus CM-abstinence improved continuous abstinence. CBT plus CM-abstinence may improve continuous abstinence compared with CBT plus CM-attendance. The impact of CBT or CBT plus CM on adverse events (safety) was unclear.

We found DBT/ACT compared with inactive/nonspecific comparators may improve point abstinence at the end of treatment. Community reinforcement was less likely to improve point abstinence when compared with CBT.46 Although we found a clinically meaningful effect of community reinforcement on continuous abstinence relative to a nonspecific comparator, the definition of continuous abstinence used was unclear and this finding is highly uncertain.54 There was no evidence that PSIs other than CBT affected completion of treatment.

For secondary outcomes at the end of treatment, we found some evidence that CBT-based interventions may reduce the quantity of cannabis used. Except for DBT/ACT, other PSIs were unlikely to reduce frequency of use. There was little evidence that CBT-based interventions could increase the duration of continuous abstinence when compared with inactive/nonspecific or active comparators.

Two trial-based economic evaluations reported higher costs for CBT and CM relative to a nonspecific comparator,61 and for MDFT compared with CBT.60 The trialists reported little difference in effectiveness, but quality of life (EQ-5D-3L) was improved for MDFT. No studies reported intensity of withdrawal, engagement in further treatment, or dropout due to adverse events.

While our findings are consistent with previous reviews of PSIs for point abstinence13,14, they are not consistent for frequency of use, where only DBT/ACT showed potential benefit in the present review. Unlike previous meta-analyses, we did not aggregate PSIs into a single intervention category but analyzed them based on shared theoretical underpinning or techniques used.62 Previous reviews either did not assess RoB14 or used earlier versions of the Cochrane RoB tool.13 We used the RoB2 tool, which focuses on bias at the outcome- and not study-level. Earlier meta-analyses also employed random-effects models but used a DerSimonian-Laird between-study variance estimator, which may be negatively biased when study sample sizes are small and heterogeneity is large. We used a more robust REML estimator.38,63 These methodological differences may also explain why we rated the certainty of the evidence base as low to very low, in contrast to a previous assessment of low to moderate certainty.13

Strengths and limitations

Our review followed a rigorous methodology. The protocol was prospectively registered and the database searches were comprehensive and recently conducted (12-Jun-2024). However, the number of studies included per meta-analysis was small (up to six), and some relative intervention effects were based on single studies. This means that our pre-planned subgroup analyses were not possible. We also excluded some publication types, such as conference abstracts, or trial registration reports lacking outcome data.

We implemented clearly defined eligibility criteria. However, the included studies used a variety of participant inclusion criteria regarding CUD, which may have contributed to the heterogeneity of the current findings. This variability may reflect changes in the diagnostic criteria over time. However, even within the same diagnostic categories, the severity of the disorder may vary and this was not consistently reported across all studies (Table 3).

To combine studies in a meta-analysis, similar interventions and comparators were grouped. However, these groupings could impact interpretation of estimated effects. For instance, the study-level effects were greater when CBT was compared with a waitlist control, and smaller when compared with a nonspecific comparator (Figures 2, 3, 5). This ‘waitlist’ effect is well-known in behavioural intervention trials.64,65 Unfortunately, our review lacked sufficient studies for appropriately powered subgroup analyses by comparator type.

Although there has been some work on outcome toolkits,66 there is no agreed core outcome set for CUD and we found considerable variability in outcome definitions and measures across all studies. Additionally, prospectively registered protocols were not available for 15 of 22 studies. As such, it is unclear whether our selected outcomes were not measured, or were measured but not reported. Abstinence was measured using self-report or urine tests, with limited details on thresholds for detecting presence of cannabinoids, and which measure had informed the outcome assessment when both were collected. The definition of ‘continuous’ abstinence also varied across studies (6-14 weeks). Using different measures and timeframes within the abstinence outcomes likely contributed to moderate between-study heterogeneity, which in turn reduces the certainty of findings.

Only two studies assessed safety outcome and both reported that no adverse events had occurred. Neither provided a definition of ‘adverse event’ or details on measurement. Reasons for participant dropout were poorly reported across all studies. In combination with substantial missing data, this was a common source of high RoB in the results. Cannabis craving was reported in one study only,41 and none reported on the intensity of withdrawal. This was unexpected considering the nature of the PSIs, some of which included training in affect management and dealing with withdrawal. For individuals with CUD, withdrawal symptoms can occur within the first week of ceasing cannabis use.3 Except for the studies of CM-abstinence,44,45,47–50,55,58 it was not clear whether or when participants were expected to stop using cannabis. Therapeutic goals may extend beyond abstinence and include reduction in cannabis use or improvement in functioning, where withdrawal symptoms may be less likely to occur.

In addition to the absence of a core outcome set, there is no established consensus for what constitutes a clinically meaningful change in outcomes. For this review, we considered a 10% increase in abstinence and completion of treatment and halving the level of cannabis use in the intervention group would represent clinically meaningful changes. However, these thresholds may be considered high, especially from a harm-reduction perspective.67 Abstinence is difficult to achieve68 and a smaller reduction in cannabis use may be meaningful for people with CUD. It is possible that some PSIs would have been considered effective if lower thresholds had been used.

Implications for research and practice

Several PSIs demonstrated clinically meaningful effects on abstinence and level of cannabis use and, in the absence of alternative treatments, it would be reasonable to suggest they are offered for CUD. However, due to the low certainty of evidence and small number of studies this recommendation for clinical practice is tentative.

To improve the robustness of the evidence base, and inform policy and practice, additional high-quality RCTs are needed. This is the same conclusion reached by Gates et al in their 2016 Cochrane review.13 Studies should be prospectively registered with published trial protocols to minimize bias from selection of the reported results. Studies should be adequately powered, and ensure the assessors are blinded, at least to the alternative intervention in case of self-reported outcomes. To reduce the impact of missing outcome data, trials should incorporate strategies to retain and follow-up participants and clearly report information concerning those who have withdrawn. Future research on core outcome sets and standardized measurement of outcomes should also involve people with CUD.

Many trials in this review were conducted over a decade ago and may not generalise to contemporary cannabis use. In recent years, there has been a surge in CUD incidence among adolescents and young adults.6 While we had insufficient data to explore whether age may moderate the effectiveness of PSIs, the specific needs of young people should be considered within treatment services. The potency of cannabis has also increased over time,69,70 and is associated with risk of CUD and more severe dependence,71,72 and may reduce treatment effectiveness. Future studies should also ensure participants from more diverse backgrounds are recruited, as the current evidence base predominantly includes white male participants. Only three studies included people with affective problems,42,43,57 and four others reported mental health co-morbidities in some participants,44,45,55,58 whereas most excluded people with severe mental health problems (e.g. suicide risk, psychosis). Mental health disorders commonly co-occur with CUD,73,74 as does nicotine dependence. Evaluation of integrated treatments may be valuable to improve outcomes for people with CUD.75,76

Such research would underpin the development of cannabis-specific, evidence-based, practice guidelines.

Conclusions

This review found that CBT, DBT/ACT and community reinforcement may be superior to inactive/nonspecific interventions for the treatment of CUD. Effectiveness and cost-effectiveness of other therapies and their combinations were less clear. The evidence for the effectiveness and safety of PSIs is of low to very low certainty. Methodologically robust trials conducted in representative samples of people with CUD are needed to inform more certain recommendations for policy and clinical practice.

ADDITIONAL INFORMATION

CRediT statement

Conceptualization: FS, SS, JS, JH, DC; Data Curation: MH, TP, KW, FS, SD, CP; Formal analysis: MH; Funding acquisition: JS, JH, DC; Investigation: MH, TP, KW, FS, DC; Methodology: MH, LAH, TPF, SS, SD, JS, JH, DC; Project administration: MH, DC; Resources: MH; Software: MH; Supervision: LAH, TPF, SS, JH, DC; Validation: MH, TP, KW; Visualization: MH, TP, KW; Writing – Original draft: MH, KW, DC; Writing – reviewing and editing: MH, TP, KW, FS, LAH, TPF, SS, JS, JH, DC.

Data availability statement

The data to support the findings of this review are available in Table 3, Figures 2-5, and Supporting Information 8-13.

SUPPORTING INFORMATION LEGENDS

Supporting Information 1. Intervention groupings (.docx)

Supporting Information 2. Operationalization of outcomes and hierarchy of preference for outcome measures and timepoints (.docx)

Supporting Information 3. Search strategies (.docx)

Supporting Information 4. Data processing (.docx)

Supporting Information 5. Additional comparisons (.docx)

Supporting Information 6. Certainty of evidence (GRADE) criteria (.docx)

Supporting Information 7. Excluded studies (.docx)

Supporting Information 8. PROGRESS-Plus characteristics (.docx)

Supporting Information 9. Risk of bias assessment (.docx)

Supporting Information 10. Results synthesis for outcomes assessed at medium and long follow-up (.docx)

Supporting Information 11. Results of sensitivity analyses (.docx)

Supporting Information 12. Summary of findings tables (.docx)

Supporting Information 13. Economic evaluation studies (.docx)

Footnotes

  • Declaration of interest: None to declare.

  • Primary funding: This review was funded by the NIHR Evidence Synthesis Programme (NIHR167862). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

REFERENCES

  1. 1.↵
    UNODC. World Drug Report 2024: United Nations publication; 2024.
  2. 2.↵
    Robinson T, Ali MU, Easterbrook B, Coronado-Montoya S, Daldegan-Bueno D, Hall W, et al. Identifying risk-thresholds for the association between frequency of cannabis use and development of cannabis use disorder: A systematic review and meta-analysis. Drug and Alcohol Dependence. 2022;238:109582.
    OpenUrlPubMed
  3. 3.↵
    American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5: American psychiatric association Washington, DC; 2013.
  4. 4.↵
    American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th ed. Arlington, VA, US: American Psychiatric Publishing, Inc.; 1994. 886 p.
  5. 5.↵
    World Health Organization (WHO). International Classification of Diseases, Eleventh Revision (ICD-11). 2019/2021.
  6. 6.↵
    Shao H, Du H, Gan Q, Ye D, Chen Z, Zhu Y, et al. Trends of the Global Burden of Disease Attributable to Cannabis Use Disorder in 204 Countries and Territories, 1990–2019: Results from the Disease Burden Study 2019. International Journal of Mental Health and Addiction. 2024;22(4):2485–507.
    OpenUrl
  7. 7.↵
    World Health Organization. The health and social effects of nonmedical cannabis use: World Health Organization; 2016 [cited 2024 Oct 18]. Available from: https://www.who.int/publications/i/item/9789241510240
  8. 8.↵
    European Monitoring Centre for Drugs and Drug Addiction. European Drug Report 2024: Trends and Developments 2024 [cited 2024 Nov 5]. Available from: https://www.emcdda.europa.eu/publications/european-drug-report/2024_en
  9. 9.↵
    Manthey J, Freeman TP, Kilian C, López-Pelayo H, Rehm J. Public health monitoring of cannabis use in Europe: prevalence of use, cannabis potency, and treatment rates. The Lancet Regional Health–Europe. 2021;10.
  10. 10.↵
    World Health Organization. Mental health gap action Programme (mhGAP) guideline for mental, neurological and substance use disorders: World Health Organization; 2023.
  11. 11.↵
    Clinical Guidelines on Drug Misuse and Dependence Update 2017 Independent Expert Working Group. Drug misuse and dependence: UK guidelines on clinical management. London: Department of Health; 2017 [cited 2024 Oct 18]. Available from: https://www.gov.uk/government/publications/drug-misuse-and-dependence-uk-guidelines-on-clinical-management
  12. 12.↵
    Schettino J, Leuschner F, Kasten L, Tossmann P, Hoch E. Treatment of cannabis-related disorders in Europe: Publications Office of the European Union; 2015.
  13. 13.↵
    Gates PJ, Sabioni P, Copeland J, Le Foll B, Gowing L. Psychosocial interventions for cannabis use disorder. Cochrane Database of Systematic Reviews. 2016(5).
  14. 14.↵
    Davis ML, Powers MB, Handelsman P, Medina JL, Zvolensky M, Smits JAJ. Behavioral therapies for treatment-seeking cannabis users: a meta-analysis of randomized controlled trials. Evaluation & the health professions. 2015;38(1):94–114.
    OpenUrlPubMed
  15. 15.
    Halladay J, Scherer J, MacKillop J, Woock R, Petker T, Linton V, et al. Brief interventions for cannabis use in emerging adults: A systematic review, meta-analysis, and evidence map. Drug and Alcohol Dependence. 2019;204:107565.
    OpenUrlCrossRefPubMed
  16. 16.
    Chatters R, Cooper K, Day E, Knight M, Lagundoye O, Wong R, et al. Psychological and psychosocial interventions for cannabis cessation in adults: A systematic review. Addiction Research & Theory. 2016;24(2):93–110.
    OpenUrl
  17. 17.↵
    Ghafouri M, Correa da Costa S, Zare Dehnavi A, Gold MS, Rummans TA. Treatments for Cannabis Use Disorder across the Lifespan: A Systematic Review. Brain Sciences. 2024;14(3):227.
    OpenUrlPubMed
  18. 18.↵
    Nielsen S, Gowing L, Sabioni P, Le Foll B. Pharmacotherapies for cannabis dependence. Cochrane Database of Systematic Reviews. 2019;1:CD008940.
    OpenUrl
  19. 19.↵
    Connor JP, Manthey J, Hall W, Stjepanović D. Effectiveness of cannabis use and cannabis use disorder interventions: a European and international data synthesis. European archives of psychiatry and clinical neuroscience. 2024:1–13.
  20. 20.↵
    Winters KC, Mader J, Budney AJ, Stanger C, Knapp AA, Walker DD. Interventions for cannabis use disorder. Current Opinion in Psychology. 2021;38:67–74.
    OpenUrlPubMed
  21. 21.↵
    Spiga F, Savovic J, Halicka M, Parkhouse T, Dawson S, Sanghera S, Higgins J, Caldwell D. Effectiveness and safety of psychosocial interventions for the treatment of cannabinoid use disorder. PROSPERO. 2024:CRD42024553382.
  22. 22.↵
    Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. British Medical Journal. 2021;372:n71.
    OpenUrlFREE Full Text
  23. 23.↵
    Noel-Storr AH, Dooley G, Wisniewski S, Glanville J, Thomas J, Cox S, et al. Cochrane Centralised Search Service showed high sensitivity identifying randomized controlled trials: a retrospective analysis. Journal of Clinical Epidemiology. 2020;127:142–50.
    OpenUrlCrossRefPubMed
  24. 24.↵
    Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan—a web and mobile app for systematic reviews. Systematic reviews. 2016;5:1–10.
    OpenUrlCrossRefPubMed
  25. 25.↵
    Laser AI [Software]. Evidence Prime; 2024 [cited 2024 Oct 18]. Available from: https://laser.ai
  26. 26.↵
    O’Neill J, Tabish H, Welch V, Petticrew M, Pottie K, Clarke M, et al. Applying an equity lens to interventions: using PROGRESS ensures consideration of socially stratifying factors to illuminate inequities in health. Journal of Clinical Epidemiology. 2014;67(1):56–64.
    OpenUrlCrossRefPubMed
  27. 27.↵
    Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. British Medical Journal. 2019;366.
  28. 28.↵
    Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ Economic Evaluation Working Party. British Medical Journal. 1996;313(7052):275.
    OpenUrlFREE Full Text
  29. 29.↵
    R Core Team. R: A language and environment for statistical computing Vienna, Austria: R Foundation for Statistical Computing; 2020 [cited 2024 Oct 18]. Available from: https://www.R-project.org/
  30. 30.
    RStudio Team. RStudio: Integrated Development for R Boston, MA: RStudio, PBC; 2020 [cited 2024 Oct 18]. Available from: http://www.rstudio.com/
  31. 31.↵
    Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. BMJ Ment Health. 2019;22(4):153–60.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    Bakbergenuly I, Hoaglin DC, Kulinskaya E. Methods for estimating between-study variance and overall effect in meta-analysis of odds ratios. Research Synthesis Methods. 2020;11(3):426–42.
    OpenUrlPubMed
  33. 33.↵
    Higgins JPT, Whitehead A. Borrowing strength from external trials in a meta-analysis. Statistics in Medicine. 1996;15(24):2733–49.
    OpenUrlCrossRefPubMedWeb of Science
  34. 34.↵
    Friedrich JO, Adhikari NKJ, Beyene J. The ratio of means method as an alternative to mean differences for analyzing continuous outcome variables in meta-analysis: A simulation study. BMC Medical Research Methodology. 2008;8(1):32.
    OpenUrlPubMed
  35. 35.
    Daly C, Anwer S, Welton NJ, Dias S, Ades AE. NICE Guidelines Technical Support Unit: Meta-Analysis of event outcomes: Guideline Methodology Document 3; 2021 [cited 2024 Oct 18]. Available from: https://www.bristol.ac.uk/media-library/sites/social-community-medicine/documents/mpes/gmd-3-events-jan2021.pdf
  36. 36.↵
    Daly C, Dias S, Welton NJ, Anwer S, Ades AE. NICE Guidelines Technical Support Unit: Meta-Analysis of continuous outcomes: Guideline Methodology Document 2; 2021 [cited 2024 Oct 18]. Available from: https://www.bristol.ac.uk/media-library/sites/social-community-medicine/documents/mpes/gmd-2-continuous-jan2021.pdf
  37. 37.↵
    Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. British Medical Journal. 2008;336(7650):924–6.
    OpenUrlFREE Full Text
  38. 38.↵
    Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al., editors. Cochrane Handbook for Systematic Reviews of Interventions. version 6.5 (updated August 2024): Cochrane; 2024.
  39. 39.↵
    NCT02102230. CBT-I for Cannabis Use. https://clinicaltrialsgov/show/NCT02102230. 2014.
  40. 40.↵
    Hoch E, Bühringer G, Pixa A, Dittmer K, Henker J, Seifert A, et al. CANDIS treatment program for cannabis use disorders: findings from a randomized multi-site translational trial. Drug and Alcohol Dependence. 2014;134:185–93.
    OpenUrlPubMed
  41. 41.↵
    Davoudi M, Allame Z, Foroughi A, Taheri AA. A pilot randomized controlled trial of dialectical behavior therapy (DBT) for reducing craving and achieving cessation in patients with marijuana use disorder: feasibility, acceptability, and appropriateness. Trends in psychiatry and psychotherapy. 2021;43(4):302–10.
    OpenUrl
  42. 42.↵
    Davoudi M, Taheri A, Foroughi A. Effectiveness of Acceptance and Commitment Therapy on Depression, Anxiety and Cessation in Marijuana Use Disorder: a Randomized Clinical Trial. International journal of behavioral sciences. 2021;15(3):194–200.
    OpenUrl
  43. 43.↵
    Buckner JD, Zvolensky MJ, Ecker AH, Schmidt NB, Lewis EM, Paulus DJ, et al. Integrated cognitive behavioral therapy for comorbid cannabis use and anxiety disorders: a pilot randomized controlled trial. Behaviour research and therapy. 2019;115:38–45.
    OpenUrlPubMed
  44. 44.↵
    Budney AJ, Higgins ST, Radonovich KJ, Novy PL. Adding voucher-based incentives to coping skills and motivational enhancement improves outcomes during treatment for marijuana dependence. Journal of Consulting and Clinical Psychology. 2000;68(6):1051–61.
    OpenUrlCrossRefPubMedWeb of Science
  45. 45.↵
    Stanger C, Budney AJ, Kamon JL, Thostensen J. A randomized trial of contingency management for adolescent marijuana abuse and dependence. Drug and Alcohol Dependence. 2009;105(3):240–7.
    OpenUrlCrossRefPubMedWeb of Science
  46. 46.↵
    Kaminer Y, Ohannessian CM, Burke RH. Adolescents with cannabis use disorders: adaptive treatment for poor responders. Addictive behaviors. 2017;70:102–6.
    OpenUrlPubMed
  47. 47.↵
    Budney AJ, Moore BA, Rocha HL, Higgins ST. Clinical trial of abstinence-based vouchers and cognitive-behavioral therapy for cannabis dependence. Journal of Consulting and Clinical Psychology. 2006;74(2):307–16.
    OpenUrlCrossRefPubMedWeb of Science
  48. 48.↵
    Kadden RM, Litt MD, Kabela-Cormier E, Petry NM. Abstinence rates following behavioral treatments for marijuana dependence. Addictive behaviors. 2007;32(6):1220–36.
    OpenUrlCrossRefPubMed
  49. 49.
    Litt MD, Kadden RM, Petry NM. Behavioral treatment for marijuana dependence: randomized trial of contingency management and self-efficacy enhancement. Addictive behaviors. 2013;38(3):1764–75.
    OpenUrlCrossRefPubMed
  50. 50.↵
    Litt MD, Kadden RM, Tennen H, Petry NM. Individualized assessment and treatment program (IATP) for cannabis use disorder: randomized controlled trial with and without contingency management. Psychology of addictive behaviors. 2020;34(1):40–51.
    OpenUrlPubMed
  51. 51.↵
    Babor TF. Brief Treatments for Cannabis Dependence: Findings From a Randomized Multisite Trial. Journal of Consulting and Clinical Psychology. 2004;72(3):455–66.
    OpenUrlCrossRefPubMedWeb of Science
  52. 52.↵
    Stephens RS, Roffman RA, Curtin L. Comparison of extended versus brief treatments for marijuana use. Journal of Consulting and Clinical Psychology. 2000;68(5):898–908.
    OpenUrlCrossRefPubMedWeb of Science
  53. 53.↵
    Stephens RS, Roffman RA, Simpson EE. Treating adult marijuana dependence: a test of the relapse prevention model. Journal of Consulting and Clinical Psychology. 1994;62(1):92–9.
    OpenUrlCrossRefPubMedWeb of Science
  54. 54.↵
    Khalily MT, Hussain B, Hallahan B, Irfan S, Ehsan N, Saghir M, et al. Indigenously adapted community reinforcement approach (ia-cra) for cannabis users: A randomized controlled trial. International Journal of Mental Health and Addiction. 2023:1–12.
  55. 55.↵
    Carroll KM, Easton CJ, Nich C, Hunkele KA, Neavins TM, Sinha R, et al. The use of contingency management and motivational/skills-building therapy to treat young adults with marijuana dependence. Journal of Consulting and Clinical Psychology. 2006;74(5):955–66.
    OpenUrlCrossRefPubMedWeb of Science
  56. 56.
    Budak FK, Akbeniz A, Erkan FM, Gultekin A, Cumurcu HB. The effect of mindfulness-based psychoeducation on negative automatic thoughts and medication adherence in individuals with cannabis use disorder: A randomized controlled trial. International Journal of Mental Health and Addiction. 2024:1–14.
  57. 57.↵
    Wolitzky-Taylor K, Glasner S, Tanner A, Ghahremani DG, London ED. Targeting maladaptive reactivity to negative affect in emerging adults with cannabis use disorder: a preliminary test and proof of concept. Behaviour research and therapy. 2022;150:104032.
    OpenUrlPubMed
  58. 58.↵
    Carroll KM, Nich C, Lapaglia DM, Peters EN, Easton CJ, Petry NM. Combining cognitive behavioral therapy and contingency management to enhance their effects in treating cannabis dependence: less can be more, more or less. Addiction. 2012;107(9):1650–9.
    OpenUrlPubMed
  59. 59.↵
    Rigter H, Henderson CE, Pelc I, Tossmann P, Phan O, Hendriks V, et al. Multidimensional family therapy lowers the rate of cannabis dependence in adolescents: a randomised controlled trial in Western European outpatient settings. Drug and Alcohol Dependence. 2013;130(1):85–93.
    OpenUrlCrossRefPubMed
  60. 60.↵
    Goorden M, van der Schee E, Hendriks VM, Hakkaart-van Roijen L. Cost-effectiveness of multidimensional family therapy compared to cognitive behavioral therapy for adolescents with a cannabis use disorder: data from a randomized controlled trial. Drug and Alcohol Dependence. 2016;162:154–61.
    OpenUrlPubMed
  61. 61.↵
    Olmstead TA, Sindelar JL, Easton CJ, Carroll KM. The cost-effectiveness of four treatments for marijuana dependence. Addiction. 2007;102(9):1443–53.
    OpenUrlCrossRefPubMedWeb of Science
  62. 62.↵
    Higgins JPT, López-López JA, Becker BJ, Davies SR, Dawson S, Grimshaw JM, et al. Synthesising quantitative evidence in systematic reviews of complex health interventions. BMJ Global Health. 2019;4(Suppl 1):e000858.
    OpenUrlAbstract/FREE Full Text
  63. 63.↵
    Langan D, Higgins JPT, Jackson D, Bowden J, Veroniki AA, Kontopantelis E, et al. A comparison of heterogeneity variance estimators in simulated random-effects meta-analyses. Research Synthesis Methods. 2019;10(1):83–98.
    OpenUrlPubMed
  64. 64.↵
    Kraiss J, Viechtbauer W, Black N, Johnston M, Hartmann-Boyce J, Eisma M, et al. Estimating the true effectiveness of smoking cessation interventions under variable comparator conditions: A systematic review and meta-regression. Addiction. 2023;118(10):1835–50.
    OpenUrlPubMed
  65. 65.↵
    de Bruin M, Black N, Javornik N, Viechtbauer W, Eisma MC, Hartman-Boyce J, et al. Underreporting of the active content of behavioural interventions: a systematic review and meta-analysis of randomised trials of smoking cessation interventions. Health Psychology Review. 2021;15(2):195–213.
    OpenUrlPubMed
  66. 66.↵
    Loflin MJE, Kiluk BD, Huestis MA, Aklin WM, Budney AJ, Carroll KM, et al. The state of clinical outcome assessments for cannabis use disorder clinical trials: A review and research agenda. Drug and Alcohol Dependence. 2020;212:107993.
    OpenUrlPubMed
  67. 67.↵
    Ritter A, Cameron J. A review of the efficacy and effectiveness of harm reduction strategies for alcohol, tobacco and illicit drugs. Drug and alcohol review. 2006;25(6):611–24.
    OpenUrlCrossRefPubMedWeb of Science
  68. 68.↵
    Volkow ND, Woodcock J, Compton WM, Throckmorton DC, Skolnick P, Hertz S, et al. Medication development in opioid addiction: Meaningful clinical end points. Science translational medicine. 2018;10(434):eaan2595.
    OpenUrlFREE Full Text
  69. 69.↵
    ElSohly MA, Mehmedic Z, Foster S, Gon C, Chandra S, Church JC. Changes in Cannabis Potency Over the Last 2 Decades (1995–2014): Analysis of Current Data in the United States. Biological Psychiatry. 2016;79(7):613–9.
    OpenUrlCrossRefPubMed
  70. 70.↵
    Freeman TP, Craft S, Wilson J, Stylianou S, ElSohly M, Di Forti M, et al. Changes in delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD) concentrations in cannabis over time: systematic review and meta-analysis. Addiction. 2021;116(5):1000–10.
    OpenUrlPubMed
  71. 71.↵
    Petrilli K, Ofori S, Hines L, Taylor G, Adams S, Freeman TP. Association of cannabis potency with mental ill health and addiction: a systematic review. The Lancet Psychiatry. 2022;9(9):736–50.
    OpenUrlPubMed
  72. 72.↵
    Freeman TP, Winstock AR. Examining the profile of high-potency cannabis and its association with severity of cannabis dependence. Psychological Medicine. 2015;45(15):3181–9.
    OpenUrlCrossRefPubMed
  73. 73.↵
    Conway KP, Compton W, Stinson FS, Grant BF. Lifetime comorbidity of DSM-IV mood and anxiety disorders and specific drug use disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of clinical Psychiatry. 2006;67(2):247–57.
    OpenUrl
  74. 74.↵
    Hasin DS, Kerridge BT, Saha TD, Huang B, Pickering R, Smith SM, et al. Prevalence and Correlates of DSM-5 Cannabis Use Disorder, 2012-2013: Findings from the National Epidemiologic Survey on Alcohol and Related Conditions–III. American Journal of Psychiatry. 2016;173(6):588–99.
    OpenUrlCrossRefPubMed
  75. 75.↵
    Hindocha C, Freeman TP, Ferris JA, Lynskey MT, Winstock AR. No Smoke without Tobacco: A Global Overview of Cannabis and Tobacco Routes of Administration and Their Association with Intention to Quit. Frontiers in Psychiatry. 2016;7.
  76. 76.↵
    Lees R, Hines LA, D’Souza DC, Stothart G, Di Forti M, Hoch E, et al. Psychosocial and pharmacological treatments for cannabis use disorder and mental health comorbidities: a narrative review. Psychological Medicine. 2021;51(3):353–64.
    OpenUrlPubMed
  77. 77.
    Montgomery L, Petry NM, Carroll KM. Moderating effects of race in clinical trial participation and outcomes among marijuana-dependent young adults. Drug and Alcohol Dependence. 2012;126(3):333–9.
    OpenUrlPubMed
  78. 78.
    Copeland J, Swift W, Rees V. Clinical profile of participants in a brief intervention program for cannabis use disorder. Journal of substance abuse treatment. 2001;20(1):45–52.
    OpenUrlCrossRefPubMedWeb of Science
  79. 79.
    Copeland J, Swift W, Roffman R, Stephens R. A randomized controlled trial of brief cognitive-behavioral interventions for cannabis use disorder. Journal of substance abuse treatment. 2001;21(2):55–64.
    OpenUrlCrossRefPubMedWeb of Science
  80. 80.
    Litt MD, Kadden RM, Kabela-Cormier E, Petry NM. Coping skills training and contingency management treatments for marijuana dependence: exploring mechanisms of behavior change. Addiction. 2008;103(4):638–48.
    OpenUrlCrossRefPubMedWeb of Science
  81. 81.
    Phan O, Henderson CE, Angelidis T, Weil P, van Toorn M, Rigter R, et al. European youth care sites serve different populations of adolescents with cannabis use disorder. Baseline and referral data from the INCANT trial. BMC Psychiatry. 2011;11:110.
    OpenUrlCrossRefPubMed
  82. 82.
    Rigter H, Pelc I, Tossmann P, Phan O, Grichting E, Hendriks V, et al. INCANT: a transnational randomized trial of multidimensional family therapy versus treatment as usual for adolescents with cannabis use disorder. BMC Psychiatry. 2010;10:28.
    OpenUrlCrossRefPubMed
  83. 83.
    Rowe C, Rigter H, Henderson C, Gantner A, Mos K, Nielsen P, et al. Implementation fidelity of Multidimensional Family Therapy in an international trial. Journal of substance abuse treatment. 2013;44(4):391–9.
    OpenUrlPubMed
  84. 84.
    Stephens RS, Wertz JS, Roffman RA. Predictors of marijuana treatment outcomes: the role of self-efficacy. Journal of substance abuse. 1993;5(4):341–53.
    OpenUrlCrossRefPubMed
Back to top
Posted November 18, 2024.
Download PDF

Supplementary Material

Data/Code
Email
Effectiveness and safety of psychosocial interventions for the treatment of cannabis use disorder: a systematic review and meta-analysis
Monika Halicka, Thomas L Parkhouse, Katie Webster, Francesca Spiga, Lindsey A Hines, Tom P Freeman, Sabina Sanghera, Sarah Dawson, Craig Paterson, Jelena Savović, Julian PT Higgins, Deborah M Caldwell
medRxiv 2024.11.18.24317475; doi: https://doi.org/10.1101/2024.11.18.24317475
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area