Treatment Implications Associated with Cannabis and Tobacco Co-Use

Abstract

Purpose of the Review.

The goal of this article is to summarize the treatment-focused literature on cannabis and tobacco co-use and the treatment implications of co-use. This review will focus on: 1) the impact of co-use on cessation outcomes, 2) compensatory use/substitution of the non-treated substance among co-users, and 3) treatment interventions to address co-use. This article will highlight the limitations to co-use captured in the literature and offer considerations and directives for co-use research and treatment moving forward.

Recent Findings.

The degree to which co-use affects cessation for a single, targeted substance remains in question, as the literature is largely mixed. Cannabis treatment trials are better equipped to answer these questions given that they do not typically exclude tobacco users. While the relationship between tobacco use and poorer cannabis outcomes appears to have some evidence, the reverse relationship (cannabis use affecting tobacco outcomes) is not consistently supported.

Summary.

The co-use of cannabis and tobacco and its impact on single substance cessation and/or compensatory substance use during cessation is generally overlooked in treatment trials, while interventions to address both substances are rare. Capturing co-use adds burden for researchers, clinicians, and participants, but is warranted given the prevalence of co-use and a rapidly changing cannabis and tobacco regulatory environment, which may further complicate co-occurring substance use. Co-users are a heterogeneous population; trials focused on co-users, in addition to better data capture and consistent terminology, will aid in an understanding of nuanced patterns of co-use critical to inform treatment interventions.

1. Introduction

Prevalence of tobacco and cannabis co-use

The use of cannabis and/or tobacco is common throughout the world. In the United States (US), rates of tobacco use (mostly via combustible cigarette smoking) have been steadily declining over the past several decades [1], though tobacco use continues to be common globally with 1.3 billion tobacco users across the world; 80% of which live in low- or middle- income countries [2]. Cannabis use is also common, with estimates of approximately 192 million past-year cannabis users globally (ages 15-64) [3]. Rates of cannabis use have been stable in most global regions, though the US has seen increased rates of cannabis use in adults [3, 4], with substantial increases occurring among adults aged 26 and over [5]. Increased cannabis use prevalence is partially due to the reduced perception of harm associated with cannabis use, potential medical benefits, and relaxed cannabis legislation [6-8]. Many individual states and countries have passed legislation to allow legal cannabis markets. Cannabis is legal for medical use in 33 US states, while recreational use is legal in 11 US states, the District of Columbia, Canada, and Uruguay.

Given the prevalence of tobacco and cannabis use separately and their traditionally common route of administration (i.e., inhaled), it is not surprising that these substances are often used together [9-18]. While overall rates of cannabis and tobacco co-use in the US appear to have increased modestly from 4.4% in 2003 to 5.2% in 2012 [19], the prevalence of daily cannabis use among daily cigarette smokers has nearly doubled from 4.9% in 2002 to 9.0% in 2014 [20]. One out of every 14 daily cigarette smokers (aged 26+) are daily cannabis users and among those aged 18-25, one out of every five daily cigarette smokers are daily cannabis users [20]. Cannabis use is common among those who use tobacco [20-23] and tobacco use is common among those who use cannabis [16, 18, 24, 25], such that in the US, greater than 60% of adult cannabis users report current tobacco use [5] in comparison to rates of 16% in the general population [1]. Though co-use is prevalent, treatment research and surveillance have not consistently captured co-use characteristics, firmly established treatment implications among co-users, or developed efficacious interventions to address co-use.

Definitions of cannabis-tobacco co-use

The co-use of cannabis and tobacco may occur in several forms and generally, research studies, national household surveys, or clinical assessments do not capture sufficient information on types of co-use. There is also a lack of consistent terminology to classify co-use patterns. Co-use may occur as: 1) simultaneous use or co-administration, which includes blunts (cigar wrappers filled with cannabis) or spliffs (joints that include cannabis and loose leaf tobacco, also referred to as joints in the UK or mulled cigarettes elsewhere), or 2) the catch-all term of concurrent use, which may include sequential use (‘chasing’ cannabis with tobacco), for substitution purposes (using one when the other is not available), or in a completely asynchronous manner, in which the individual uses both substances but with no temporal proximity or relationship in use. In a recent Addiction Opinion and Debate, we have argued for consistent terminology in the field to clarify patterns of co-use, particularly as it pertains to population-level data and the unknown harms associated with co-use due to insufficient characterization [26].

Harms associated with co-use

Cannabis and tobacco co-use is associated with harm in the following areas: 1) greater prevalence of psychiatric and psychosocial problems in both youth and adults [27, 28], 2) additive health risk [29], and 3) lower self-reported ratings of health among co-users [30]. The initiation and use of one substance has been associated with increased risk of initiation of the other substance, with directionality of increased risk occurring in both directions (i.e., cannabis → tobacco and tobacco → cannabis) [31-34]. Beyond initiation, co-use may contribute to more severe substance use disorder (SUD) presentation. The use of cannabis has been associated with increased risk and greater levels of nicotine dependence [35-37] and cigarette smoking has been associated with severity of cannabis dependence [38, 39]. Further, the rapidly changing cannabis and tobacco regulatory landscape has the potential to impact co-use as novel delivery methods are available. Electronic nicotine delivery systems (ENDS) have dramatically changed nicotine and tobacco control efforts and these devices have implications for co-use [26, 40], while cannabis legalization has led to a thriving marketplace with a wide diversity of products [41, 42], including an increase in products used to vaporize cannabis among adolescent and adult cannabis users [43-45].

There are important treatment-related concerns specific to co-use, which is the focus of the current article. Though the prevalence of co-use is high, there is little consensus regarding treatment recommendations and cessation strategies tailored for co-users. There are critical gaps in the literature regarding co-use as it impacts treatment and considerations in the study of both cannabis and tobacco cessation. Therefore, the goal of this article is to review the treatment-focused literature that explores cannabis and tobacco co-use and assess the implications of co-use on treatment outcomes. Specifically, we will focus on the literature exploring: 1) the impact of co-use on cessation outcomes, 2) compensatory use/substitution of the non-treated substance among co-users, and 3) treatment interventions to address co-use. In addition, this article will offer considerations and directives for co-use research and treatment moving forward.

2. Treatment Implications of Co-Use

Impact of co-use on cessation outcomes

The role of co-use in cessation outcomes has been evaluated thus far through secondary analyses or epidemiological studies using community samples. While this literature provides useful information regarding cessation outcomes among co-users, there are limitations to drawing conclusions based on these studies alone, particularly when there is heterogeneity in study samples, collection of co-use, definitions of co-use, and mixed results regarding cessation outcomes. Available data to better understand this relationship are limited, resulting in gaps in the literature.

A recent systematic review and meta-analysis evaluated the impact of interventions that were designed to address cannabis, tobacco, co-use and/or poly-substance use among co-users of cannabis and tobacco (N=20) [46]. The conclusions of the review were that: 1) interventions showed weak evidence of an effect on cannabis cessation, but a significant effect on cannabis reduction; 2) there was no clear effect on tobacco cessation or reduction, and 3) co-use interventions appear feasible. The current review will not replicate the timely and important contribution of Walsh and colleagues but will review the co-use literature that has focused specifically on cessation outcomes through both clinical trials and cross-sectional analyses published in the past 10 years. We further narrowed our focus to studies that included behavioral outcomes, rather than necessarily self-reported outcomes (i.e., quit interest, motivation, etc.), though we acknowledge the important role of motivation, interest, and cessation intentions.

Cannabis use impacting tobacco cessation outcomes

Studies focused on tobacco cessation or treatment outcomes among co-users are shown in Table 1 (N=14). Key details of each study are listed, in addition to any evidence suggesting an impact of co-use (yes or no). Positive (+) or negative (−) impacts of co-use are also shown in the table (no impact/neutral or no change is represented as positive [+]), with qualifiers and/or clarification regarding study results. Some evidence suggests that cannabis co-use has been associated with an adverse impact on tobacco outcomes in the form of lower rates of tobacco cessation, an increased risk of relapse to smoking, faster time to lapse during a quit attempt, and has been prospectively associated with smoking amount and frequency (n=9 of 13 studies) [25, 47-54]. However, other studies (n=7 of 13 studies) found no evidence of an adverse impact of cannabis on tobacco outcomes [54-59]. Notably, several studies yielded results that support both a positive and negative impact of cannabis co-use, which varied in most cases by tobacco outcome (i.e., abstinence, time to lapse, and relapse).

Table 1.

Treatment-related studies evaluating the impact of co-use on cessation, compensatory use/substitution, or co-use interventions published in the last 10 years.

Tobacco Studies: Cannabis use/history → affecting tobacco cessation/outcomes
Study	Outcome Category	Study Design	Data Source	Study Population	Impact on Co-use	Qualifiers or Clarification
Haskins et al. (2010) [47]	Cessation	Cross-sectional baseline analysis	Parent trial	Pregnant, Latina smokers	Yes (−)	Lower odds of cessation
Metrik et al. (2011) [55]	Cessation	Secondary analysis	Parent tobacco trial	Heavy drinking smokers (ages 18+)	No (+)	No effect on smoking outcomes or relapse
	Substitution	-	-	-	Yes (+)	⇓ Cannabis use over the trial
Hendricks et al. (2012) [56]	Cessation	Secondary analysis	Three tobacco treatment clinical trials	Adult smokers (18+)	No (+)	No effect on abstinence
Leyro et al. (2015) [57]	Cessation	Secondary analysis	Parent tobacco treatment trial	Adult smokers (18+)	Yes (+)	⇑ Odds of achieving abstinence
Rabin et al. (2016) [58]	Cessation	Secondary analysis	Tobacco treatment trial	Adults smokers (age 18-65)	No (+)	Study enrolled cannabis co-users who did not meet for a cannabis use disorder
Pacek et al. (2016) [59]	Substitution	Secondary analysis	Parent RCT	Adult smokers (18+)	No (+)	Low nicotine content cigarettes did not affect cannabis use (no compensatory use)
Schauer et al. (2017) [25]	Cessation	Cross-sectional analysis	National data set (NSDUH; US)	Adult ever tobacco users (18+)	Yes (−)	⇓ Sustained abstinence
Weinberger et al. (2018) [48]	Cessation, Relapse	2-wave analysis	National data set (NESARC; US)	Adult smokers (18+)	Yes (−)	⇓ Smoking cessation; ⇑smoking relapse
El-Khoury et al. (2018) [50]	Cessation	Cross-sectional analysis	National data set (France)	Adult smokers (ages 18-64)	Yes (−)	⇓ Successful cessation
Vogel et al. (2018) [49]	Cessation	Secondary analysis	Parent RCT	Young adult smokers (ages 18-25)	Yes (−)	Impact on abstinence outcomes
Rogers et al. (2020) [51]	Cessation, lapse, relapse	Secondary analysis	Parent RCT	Adult smokers	Yes (− and +) (18-65)	Impact on abstinence, time to lapse, but not to relapse
McClure et al. (2020) [52]	Cessation	Secondary analysis	Parent RCT	Youth smokers (ages 14-21)	Yes (−)	Impact on tobacco cessation was pronounced among males and more frequent cannabis users
Voci et al. (2020) [53]	Cessation	Cross-sectional analysis	Primary care-based smoking cessation program	All primary care smokers (any age)	Yes (−)	Recreational cannabis use was associated with poorer cessation outcomes, but medical cannabis use was not
Britton et al. (2020) [54]	Cessation	Secondary analysis	Prospective study of couples (self-guided quit attempt)	Couples with one smoking partner (ages 18-55)	No (+)	Past-year smoker or partner cannabis use did not affect abstinence (trend-level for smoker, but not significant)
	Tobacco use	-	-	-	Yes (−)	Partner and smoker cannabis use associated with smoking and ⇑ CPD
Cannabis Studies: Tobacco use/history → affecting cannabis cessation/outcomes
Study	Outcome Category	Study Design	Data Source	Study Population	Impact on Co-use	Qualifiers or Clarification
Allsop et al. (2014) [60]	Cessation	Secondary analysis	Prospective parent study	Non-treatment seeking, dependent cannabis users adults	Yes (−)	Using less at baseline had ⇑ cannabis abstinence
	Substitution	-	-	-	Yes (−)	⇑ in tobacco use during cannabis abstinence
McClure et al. (2014) [61]	Cessation	Secondary analysis	Parent RCT	Cannabis-dependent youth (ages 15-21)	Yes (−)	Smokers had slightly lower rates of cannabis abstinence (NS); Among placebo → trend-level ⇓ abstinence (NS)
	Substitution	-	-	-	No (+)	No CPD changes
Gray et al. (2017) [62]	Cessation	Cannabis cessation RCT	-	Cannabis-dependent adults (ages 18-50)	Yes (−)	Smoking status was associated with worse cannabis outcomes; Tobacco smoking status was a stratification variable at randomization and included in final models
McClure et al. (2018) [65]	Cessation	Secondary analysis	Parent RCT	Cannabis-dependent adults (ages 18-50)	No (+)	ND did not affect cannabis cessation
	Substitution	-	-	-	Yes (+ and −)	Active treatment: lower ND resulted in ⇓ smoking; higher ND resulted in ⇑ smoking
Peters & Hughes (2010) [91]	Substitution	Within-subject cannabis abstinence paradigm	-	Daily, non-treatment seeking cannabis users with moderate alcohol use (age 18+)	No (+)	No CPD changes during cannabis abstinence
Rabin et al. (2018) [63]	Substitution	28 day cannabis abstinence paradigm	-	Cannabis dependent males (ages 18-55) with schizophrenia or non-psychiatric controls	Yes (−)	Transient tobacco substitution for patients w/ schizophrenia, while no changes in CPD were observed in controls
Cannabis and Tobacco Co-Use Interventions
Study	Study Design		Treatment	Results
Hill et al. (2013) [92]	Pilot feasibility, open label trial		CBT and nicotine patch	Treatment reduced tobacco use, but not cannabis; no compensatory cannabis use; Generally acceptable
Becker et al. (2015) [93]	Pilot feasibility study		Group cessation program	Treatment was feasible and acceptable, with preliminary efficacy on tobacco and cannabis use, as well as positive impact on other outcomes (drinking, depression, anxiety)
Lee et al. (2015) [94]	Pilot efficacy study		Computer-assisted behavioral treatment (for tobacco and cannabis, including MET, CBT, and CM) and NRT	Tobacco treatment during CUD treatment was feasible, has positive impact on tobacco use without adversely affecting cannabis use
Beckham et al. (2018) [95]	Pilot feasibility study		Mobile CM	Treatment is feasible and acceptable
Adams et al. (2018) [82]	Pilot RCT		Varenicline and counseling	Treatment was tolerated and may reduce cannabis craving and tobacco and cannabis use
Lee et al. (2019) [88]	Controlled trial		Sequential or simultaneous tobacco cessation in cannabis treatment context; web-based counseling and NRT	Acceptable to address tobacco use during CUD treatment, but did not lead to tobacco cessation
Freeman et al. (2020) [96]	Cannabis cessation pharmacotherapy RCT (Phase 2a)		Cannabidiol (treatment targeted cannabis use)	Study did not specifically target co-use, but participants were cannabis users (ages 16-60) who met DSM-5 criteria for CUD and reported mixing tobacco in their cannabis (joints in the UK); Primary outcomes were focused on cannabis, secondary outcomes included tobacco use; Cannabidiol reduced cannabis use and number of cigarette smoked

ND=nicotine dependence; NRT=nicotine replacement therapy; CM=contingency management; CBT=cognitive behavioral therapy; MET=motivational enhancement therapy; NS=not statistically significant; CPD=cigarettes per day; DSM-5=Diagnostic and Statistical Manual of Mental Disorders; CUD=cannabis use disorder

Many recent studies have important caveats. For example, one study found an adverse impact of cannabis on tobacco outcomes, but this was more pronounced in male participants compared to females and those with more frequent cannabis use [52]. Another study differentiated recreational and medical cannabis co-users and found that medical cannabis use did not consistently impact tobacco outcomes, while recreational cannabis use was associated with poorer tobacco cessation outcomes [53]. From this treatment-focused literature, it can be concluded that: 1) there are limitations to the current literature, such as methodological variation, lack of biochemical verification to confirm cannabis use status and severity, and variations in the study samples used, and 2) there are likely relationship-modifying factors (e.g., gender, reason for cannabis use) which add nuance to this association and are not currently being captured.

Tobacco use impacting cannabis cessation outcomes

We also explored co-use impact in the opposite direction (i.e., tobacco use affecting cannabis cessation and outcomes). We reviewed treatment-focused articles exploring co-use specifically with the emphasis on cannabis outcomes (Table 1; N=6). Generally, this literature has shown that tobacco use has an adverse impact on cannabis cessation outcomes (n=4 out of 6 studies) [60-63], which is consistent with previous reviews [64]. Co-users who present to cannabis treatment have been shown to have more severe cannabis use characteristics [65] and demonstrate greater cannabis withdrawal symptoms during abstinence [66], potentially contributing to an adverse impact of co-use on cessation outcomes. This literature is similarly limited as in the tobacco cessation literature discussed above as there is variability in the collection of co-use, use severity, and lack of biochemical verification. Further, the use of non-combustible methods of tobacco/nicotine (e.g., ENDS) has increased substantially over the past several years and there is little data investigating the influence of ENDS use or poly-tobacco use on cannabis cessation outcomes currently.

Compensatory use/substitution of the non-treated substance among co-users

Another treatment concern for co-users is the issue of compensatory use (i.e., increased use) or substitution of the other, non-treated substance, during cessation. Whereas some tobacco cessation trials may exclude co-users of cannabis, particularly those with more severe use patterns, it is uncommon for cannabis trials to exclude tobacco users from participation. Cannabis cessation studies are therefore better equipped to report on compensatory tobacco use/substitution during cannabis cessation, if those data are collected.

Among co-users, evidence of compensatory tobacco use during cannabis cessation and compensatory cannabis use during tobacco cessation has been mixed (Table 1), with no consistent pattern emerging. Similar to the cessation literature discussed above, there is likely nuance in compensatory use, variation among co-users, and challenges in the collection and quantification of cannabis use [67]. A recent study among co-users with schizophrenia found that compensatory use of tobacco was most pronounced in the first seven days of cannabis abstinence, but returned to baseline levels by Day 28 of abstinence [63]. Notably, the same pattern of increased tobacco use during cannabis abstinence was not observed in a comparison group of non-psychiatric controls. This study suggests that the compensatory relationship between these two substances may be transient during abstinence and complicated by the presence of a psychiatric illness. Granular data to detect compensatory use of the non-treated substance are generally not collected and therefore not available to draw conclusions. Compensatory use may represent an unintended adverse consequence of successful cessation for either substance.

Treatment interventions to address co-use

To date, six studies have been published evaluating co-use treatment interventions and one additional study published that enrolled cannabis-tobacco co-users but was focused on cannabis cessation (Table 1). Most have been pilot/feasibility trials and have used various forms of treatment interventions (i.e., pharmacotherapy, contingency management, etc.). These studies have shown that co-use interventions are feasible and acceptable, treatments targeting co-use do not adversely impact cannabis or tobacco outcomes, and preliminarily results show positive outcomes for tobacco and/or cannabis use, though mostly in the form of reduced use, rather than cessation. Fully-powered, randomized controlled trials will need to be conducted to establish cessation efficacy and optimal treatment modality for co-use interventions.

3. Treatment Considerations

Co-use issues needing further consideration

Available data to better understand the relationship of tobacco and cannabis co-use and how co-use affects cessation and compensatory use are limited. Co-users of tobacco and cannabis are a heterogeneous group and literature to date has failed to differentiate meaningful groups of co-users. Certain factors that may modulate the relationship between co-use of tobacco and cannabis should be considered, such as sex/gender, age, culture, as well as the presence of a psychiatric illness. Accordingly, it may be oversimplified to say that cannabis co-users have a more difficult time with tobacco cessation, or vice versa, and cessation strategies may need to be more nuanced and personalized. There are several important treatment considerations for co-users that should be addressed moving forward.

Excluding co-users from trials

It is generally standard practice in tobacco cessation trials to exclude for concurrent SUDs, which would eliminate most regular cannabis users. Secondary analyses of co-users in tobacco trials are limited by small sample sizes and may not be representative of the co-using population. If cannabis users are included, they are likely infrequent users or are inaccurately reporting use. Broad spectra of co-using populations are not specifically recruited and comparisons to tobacco only controls are not always possible. As such, severity of cannabis use may be an important variable that predicts treatment response, but current tobacco trials are not equipped to answer that question. For example, in a study listed above [58], cannabis co-use was present in 15% of the sample and participants did not meet clinical thresholds for cannabis use disorder. That study found that cannabis use did not impact tobacco cessation, which could be due to the level of cannabis use severity, which is consistent with recent youth data [52]. Future trials may consider broadening inclusion criteria to allow for co-users to be enrolled, or studies should specifically focus on recruiting a co-using population.

Insufficient collection of co-use data

When co-users are enrolled in treatment trials, assessment of cannabis or tobacco use may not be collected with sufficient granularity to detect changes in use during treatment. The collection of tobacco use during trials is somewhat straightforward among combustible tobacco users (i.e., cigarettes per day). This data capture becomes complicated when participants are using ENDS, which vary by device type, nicotine concentration, etc. making quantification challenging. Similarly, the quantification of cannabis use can be time-intensive, costly, and notoriously challenging when self-reported [67]. When co-use outcomes are not the primary outcome of the trial, it is understandable that studies have limited capacity to collect these data. Finding solutions to these data collection barriers is challenging. Our group has attempted to circumvent these barriers by implementing mobile daily diaries to better capture substance use in single-substance treatment trials [68] and relying more heavily on ambulatory assessment as part of trials [69].

Related nature of cannabis and tobacco use

Cannabis and tobacco are closely intertwined and there is not consensus in the field regarding the underlying mechanisms contributing to co-use. Several hypothesized mechanisms have been offered to explain high rates of co-use among cannabis and tobacco [70-72]. In addition to considering the treatment implications of co-use, the underlying mechanisms are critical for the development of co-use treatment interventions to increase the likelihood of sustained abstinence. The related nature of these substances and how precisely they are used together may vary across individuals and geographic regions [73], which requires additional attention. In an online survey, co-users self-reported individual differences in their perceived relatedness between substances and variation in temporal patterns of co-use [74]. Using ecological momentary assessment (EMA), another study found variability in temporal associations among co-users [75]. Patterns of co-use may also have implications for other domains. One study found that those who co-used in the same occasion reported heavier substance use and more problematic behaviors [76].

The related nature of cannabis and tobacco may also contribute to compensatory use/substitution. Concerns regarding compensatory use are further bolstered when considering withdrawal among co-users engaging in cessation and how those symptoms are managed. Human laboratory models of cannabis relapse (not reviewed here) found that co-users had worse outcomes and tobacco use predicted cannabis relapse through laboratory proxies [77]. Tobacco has also been self-reported as a way to manage cannabis withdrawal [78]. In an online survey, 62% of co-users self-reported an increase in their tobacco use during past cannabis cessation attempts, and 50% reported an increase in cannabis use during past tobacco cessation attempts [79].

Rapidly changing regulatory environment

Another issue is the impact that tobacco and cannabis regulation is exerting on co-use. The cannabis regulatory landscape is constantly evolving in the US, and its impact on co-use is still unknown. Typically, cannabis and tobacco have shared a similar route of administration, potentially contributing to their prevalent co-use [10]. Due to changing legislation and commercial availability, there is now massive variation in cannabis products and methods of administration. Edibles, concentrates (e.g., wax, dabs), and vaporizers are increasingly popular and allow for cannabis use through administration methods that do not involve combustion, potentially distinguishing them from tobacco. Indeed, it has been suggested that the uptake of cannabis vaporizers may lead to lower rates of cannabis-tobacco co-use [80]. Further, nicotine and tobacco products are rapidly changing, and products and terminology are often used together and interchangeably. Specifically, the use of cannabis in ENDS devices is problematic and becoming more common [40] and has muddled population-level data and co-use estimates.

How best to treat co-use or a co-using population

The eventual goal for treatment researchers and clinicians is to develop and implement interventions for co-users that maximize their likelihood of achieving cessation, which may include single or multi-substance cessation. How those interventions are developed and what form they take are still very much up for debate and requires further data to inform treatment development. Importantly, given that there remains no approved pharmacotherapy to treat cannabis use disorder poses an additional challenge in addressing co-use. Below, we offer directives for future research and considerations in the development of treatment strategies for co-users.

Novel and tailored co-use treatment interventions

Many co-use interventions to date (Table 1) have used a combination of pharmacotherapy, in addition to psychosocial, counseling, or behavioral strategies to support cessation. Pharmacotherapy development and evaluation is one promising area for co-use interventions. While there is currently no approved pharmacotherapy to treat cannabis use disorder, there are several evidence-based and approved pharmacotherapy options available for tobacco users. Some work has been conducted using tobacco cessation pharmacotherapies as a treatment for cannabis use. A recent study found that nicotine patch may alleviate negative affect-related cannabis withdrawal symptoms among those who were not heavy tobacco users [81], suggesting the potential use of nicotine replacement as a pharmacotherapy for co-users. One co-use treatment study included in this review [82] explored varenicline for the treatment of both tobacco and cannabis use, while a human laboratory study with co-users evaluated the combination of varenicline and nabilone on tobacco and cannabis outcomes [83]. Additional trials are ongoing to evaluate varenicline for cannabis use disorder in adults (NCT02892110, NCT03980561). The use of tobacco cessation pharmacotherapy to treat cannabis use and/or co-use is a compelling future direction to explore.

Psychosocial interventions, either combined with pharmacotherapy or alone, also hold promise in the treatment of co-use. Co-use interventions thus far have utilized a number of psychosocial or behavioral platforms to promote dual cessation, including; contingency management, group and/or individual counseling, cognitive behavioral therapy, and/or motivational enhancement therapy delivered either in-person or remotely (Table 1). While these interventions have shown early promise, future treatments may benefit from integrating personalized or tailored approaches to address the processes and mechanisms underlying co-use. For example, one recent study in adolescents suggested that treatment strategies for youth who co-use should focus on impulsivity and sensitivity to reward to modify use of both substances [84]. This suggests that targeting common processes with treatment would result in decreased use of both substances. Additionally, another study suggested that certain cannabis use motives likely play a role in co-use and the ability to successfully quit using tobacco among co-users [85]. That study further highlights the nuance of co-use and individual differences that may be important for tailored treatment. There may also be profiles of co-use and compensatory use that emerge based on certain characteristics (e.g., nicotine dependence, types and patterns of co-use, etc.) that should be addressed in a personalized manner during treatment.

The development of future interventions for co-users may require enhanced and/or stepped-care for cessation support based on certain predictive substance-using characteristics (as discussed above), as well as demographic and psychiatric characteristics. It is also important to consider the delivery method of these interventions and how their reach could be improved and expanded through remote and/or digital methods of delivery. A recent review article summarized assessment and intervention targeting co-use, specifically with the use of digital methods, such as ecological momentary assessment (EMA) or interactive voice response (IVR) systems [86]. These methods hold the potential to improve data capture and could augment data collection as part of in-person trials to better characterize and treat co-use.

Quit interest

Critical to the issue of co-use treatment and treating a co-using population is quit interest and risk perception. While co-use interventions discussed here have been feasible and acceptable to co-users, it is unclear if the majority of co-users want to quit using both substances. Particularly as medical cannabis use increases and the perception of harm associated with cannabis decreases [6, 7, 87], there are likely to be more co-users who do not have interest in abstaining from cannabis, even if they are motivated to quit using tobacco. Data from an online survey of co-users found that interest in quitting tobacco was moderate to high (7.1±2.9; 10-point scale [1-10]), while interest in quitting cannabis was low (2.4±2.3; [79]. Further, participants expressed a preference to quit smoking cigarettes first, and then address cannabis cessation, rather than undertake simultaneous cessation. Aside from patient preference, it is not yet clear if simultaneous or sequential treatment approaches may differentially impact substance use outcomes [88].

Given that co-users may not be immediately interested or motivated to abstain from cannabis or may be using cannabis medically, future treatment trials may take on a harm reduction approach for cannabis, while treating tobacco use. Tobacco users who are motivated to quit may encounter additional barriers to abstinence due to their cannabis use (e.g., in the case of simultaneous users), in which case, it should be standard practice to address cannabis concurrently with tobacco treatment. If cannabis does not impact tobacco cessation, it may be unnecessary to encourage cessation, especially if the motivation to quit using cannabis is low. Future studies may focus on cannabis harm reduction as part of tobacco cessation, but without abstinence from cannabis as the goal. Tobacco interventions among co-users may encourage lower risk cannabis use [89], could address the medical use of cannabis to assess for alternative strategies, or stress reduction, rather than abstinence goals [90]. Further, interventions could address the relationship between substances and work to decouple co-use to avoid shared vulnerability, increased craving, etc. that may contribute to use and relapse. It has been suggested that non-combustible forms of cannabis use may disentangle the relationship between cannabis and tobacco co-use and may lead to reduced rates of tobacco use [80], so harm reduction work may focus on switching to non-combustible methods of use, potentially as a step towards cessation.

4. Summary and Conclusions

The co-use of cannabis and tobacco is common across the world and its impact on single substance cessation and/or compensatory substance use during cessation is generally overlooked in treatment trials. Interventions to treat co-use are rare, despite high rates of co-use and co-users already being enrolled in treatment trials. Capturing co-use adds burden for researchers, clinicians and participants, but is warranted given co-use prevalence and a rapidly changing cannabis and tobacco regulatory environment, which may further complicate the relationship of co-occurring substance use. The literature thus far is mixed and while co-use treatment is needed, the degree to which co-use affects treatment success for a single, targeted substance remains in question. More work is needed to better understand the underlying mechanisms of co-use and to inform treatment interventions that are either tailored specifically to the needs and preferences of co-users or are focused on cessation from both substances. Co-users are a heterogeneous population and trials focused on this population, in addition to better data capture and consistent terminology, will aid in an understanding of nuanced patterns of co-use critical to inform treatment interventions.