Evaluating cannabis–nicotine substitution and demand during nicotine cessation treatment in young adults: protocol for a prospective, remote 12-week nicotine cessation trial in the USA

Abstract

Introduction

The co-use of nicotine/tobacco products and cannabis among young adults is prevalent, yet little is known regarding the treatment impact of co-use and the underlying relationship between substances. No prospective treatment studies to date have evaluated the impact of cannabis use on nicotine cessation specifically among young adults, with an emphasis on dual-use substance patterns predicting outcomes. The overall goal of this protocol is to characterise the relationship between nicotine and cannabis and evaluate how patterns of co-use are associated with nicotine cessation outcomes through a behavioural economics framework to examine how cannabis co-use may be associated with nicotine cessation. This study protocol describes an ongoing, prospective nicotine cessation trial being conducted among young adults who co-use nicotine (vaping and/or cigarettes) and cannabis products to: (1) evaluate the association between behavioural economically derived measures of substance substitutability on end of treatment (EOT) nicotine abstinence, (2) examine whether changes in nicotine use are associated with changes in cannabis demand and use and (3) assess reciprocal prospective relationships between patterns of nicotine and cannabis use during treatment.

Methods and analysis

This is a fully remote, prospective, 12-week nicotine cessation trial among young adults (ages 18–25; target N=350) across the USA who co-use nicotine and cannabis. Eligible participants vape nicotine and/or smoke cigarettes regularly (20+ days in the past 30) and use cannabis on 10+ out of the past 30 days. Participants receive comprehensive nicotine cessation treatment, which includes digital contingency management to reinforce nicotine abstinence, brief counselling and SMS-text based cessation support whereas cannabis use is monitored but not treated. Biochemical verification (through oral fluid samples) and self-reports (mobile daily diaries) of substance use are collected longitudinally. Study enrolment began in May 2024 and is expected to conclude in September 2027.

Aim 1 (primary outcome) analyses will evaluate whether baseline behavioural economic indices of nicotine–cannabis substitutability predict EOT nicotine abstinence using logistic regression models. Aim 2 (exploratory outcome) analyses will use generalised linear mixed-effects models to examine within-person associations between changes in nicotine use (abstinence, reduction in use, nicotine withdrawal) and cannabis demand and use over time, and whether these associations are moderated by time-varying substitutability. Cross-lagged panel models will be used for Aim 3 (exploratory outcome) to assess reciprocal prospective relationships between nicotine and cannabis use during treatment.

Ethics and dissemination

This study has received Institutional Review Board approval from the Medical University of South Carolina (reference ID: Pro00131124). All participants provide informed consent prior to study procedures. Findings will be disseminated through peer-reviewed publications, conference presentations and data sharing in accordance with National Institutes of Health and institutional guidelines. The manuscript reports protocol components according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist.

Trial registration number

NCT06239350.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

Strengths:

• Multimethod verification of substance use using both biochemical and self-report methods to verify substance use and abstinence, enhancing the validity of outcome measurements and rigour.

• Behavioural economic measures (purchase tasks, cross-commodity demand) allow for nuanced assessment of nicotine–cannabis substitution, complementarity and independence during a nicotine cessation trial.

• Comprehensive, multicomponent nicotine cessation intervention combining digital contingency management, counselling, hypothetical purchase tasks delivered through an online experimental marketplace and evidence-based SMS programmes, all being leveraged to evaluate the role of cannabis co-use on nicotine cessation outcomes in young adults.

• Decentralised trial design with all remote procedures allows for the enrolment of study participants across the country, ensuring a representative sample of young adults (ages 18–25) co-using nicotine and cannabis.

Limitations:

• Cannabis use is not directly targeted within the intervention, which may limit generalisability to individuals seeking treatment for nicotine–cannabis co-use and may represent an unmet treatment need for those seeking to reduce or quit cannabis during the study.

Introduction

Tobacco use rates have been declining in the USA among nearly every age group.^{1 2} This encouraging trend is overshadowed by the high rates of nicotine vaping (ie, electronic cigarettes (e-cigarettes)) and polyproduct nicotine/tobacco use among young adults (ages 18–25).^{3 4} Enhancing treatment strategies for nicotine/tobacco use in this age group is critical, particularly since combustible tobacco interventions yield only modest outcomes and nicotine vaping interventions are still in the early stages of evaluation.⁵,⁹ Of relevance is the co-use (defined as use of both substances in the same individual) of cannabis among those who use nicotine products, which is exceedingly common among young adults¹⁰,¹² as well as adults¹³,¹⁵ and may affect tobacco/nicotine cessation efforts.¹⁶,¹⁸ Nicotine and cannabis co-use is associated with public health burden in the form of greater prevalence of psychiatric and psychosocial problems,^{10 14 19} additive health risk, increased toxicant exposure and respiratory symptoms,²⁰,²³ and lower self-reported ratings of health.²⁴ There are also harms associated with co-use specific to substance initiation, dependence and treatment outcomes.²⁵ Cannabis use has been associated with an increased risk of nicotine dependence and greater nicotine dependence among those who co-use,²⁶,²⁹ and greater risk of multiple tobacco product use in young adults,³⁰ suggesting an adverse relationship between cannabis use and nicotine/tobacco cessation. Yet, little is known regarding the underlying relationships between substances and how cannabis and nicotine interact, particularly during a quit attempt.

Assessing the relationship between nicotine and cannabis is important during a quit attempt, as there appear to be treatment-related harms associated with co-use. These harms have been demonstrated in the form of worse tobacco treatment outcomes among those who co-use cannabis, mostly in adult samples, though this literature has been mixed^{25 31} and generally limited by the following factors: (1) studies have typically assessed cessation from cigarettes only, rather than assessing polynicotine cessation, including vaping, (2) studies have methodological limitations (eg, lack of biochemical verification, variations in samples) and (3) to date, three tobacco/nicotine treatment trials have assessed the impact of cannabis co-use among young adults,^{12 17 32} with mixed results, indicating that this is a priority population for the study of co-use. Further, these studies do not assess individual differences in co-use or the underlying relationship between substances, which may be associated with differential tobacco outcomes among individuals who co-use. Given the mixed literature, it is possible that there is a subset of individuals that co-use who have more difficulty achieving cessation than others, but no data on which individuals experience difficulty achieving and maintaining abstinence exist. To date, co-use patterns have not been well-characterised, and prior research has failed to capture and quantify important heterogeneity among co-using young adults and the impact of co-use differences on treatment outcomes.

Behavioural economics provides a framework with which to study an individual’s dual substance use patterns.³³,³⁶ These measures provide a quantitative metric of demand for a substance (abuse liability) and reveal the degree to which two substances serve as substitutes for an individual. In behavioural economics, two commodities can be understood as falling into one of three categories: substitutes, complements or independents. When two commodities substitute for one another, decreases in the consumption of one (eg, nicotine) result in comparable increases in consumption of the other (eg, cannabis), whereas if they served as complements, changes in consumption of both would occur in the same direction (eg, if nicotine use decreases, so does cannabis). Finally, commodities that are independent from each other do not result in any consistent changes when consumption of one is impacted.^{33 37 38} Further, the relationship between nicotine and cannabis and the nature and magnitude of the association between co-use and nicotine outcomes may be most evident during treatment, such that nicotine abstinence, reduction, withdrawal or other processes may change cannabis use and demand, which has not been evaluated in any prior studies.

The objectives of the current study are to characterise and evaluate nicotine–cannabis co-use patterns through a behavioural economics framework during nicotine cessation treatment among young adults. This trial protocol describes a remote, prospective, 12-week nicotine cessation clinical trial conducted among young adults (ages 18–25; target N=350) from across the USA who co-use nicotine and cannabis products regularly. Nicotine cessation treatment (for vaping, smoking or dual use) is provided for 12 weeks, and although cannabis use is not clinically addressed it is monitored throughout. The aims of this study are to: (1) evaluate the association between behavioural economically derived measures of substance substitutability and end of treatment (EOT) nicotine abstinence (Aim 1, primary outcome); (2) examine whether nicotine abstinence, reduction and/or nicotine withdrawal observed during treatment (a) is associated with co-occurring changes in cannabis demand and use and (b) if substance substitutability modifies this relationship (Aim 2, exploratory outcome); and (3) assess the reciprocal prospective relationship between patterns of nicotine and cannabis use during nicotine treatment (Aim 3, exploratory outcome). The overall hypothesis is that the degree of substance substitutability at the individual participant level will be differentially associated with nicotine cessation and cannabis use changes during nicotine treatment. Specifically, we hypothesise that participants with greater substitutability patterns would have lower rates of nicotine abstinence compared with participants with patterns indicating a complementary relationship, suggesting patterns consistent with less favourable cessation outcomes among some individuals who co-use. Additionally, we hypothesise that the relationship between nicotine use changes (abstinence, reduction) and/or withdrawal during treatment and changes in cannabis demand and use would be moderated by time varying changes in substitutability. The objectives of this study will inform both the potential benefits of nicotine cessation interventions and potential unintended consequences, including changes in cannabis use and related outcomes during nicotine cessation treatment.

Methods and analysis

Research design

Study overview

This study is a 12-week nicotine cessation trial among young adults (ages 18–25; N=350) from across the USA who co-use nicotine and cannabis. Participants are included who vape nicotine and/or smoke cigarettes regularly (20+ days in the past 30) and use cannabis (any method) 10+ days out of the past 30. All participants receive a nicotine cessation intervention, including digital contingency management (CM) for nicotine abstinence, brief weekly counselling focused on nicotine cessation and recommendations to use evidence-based SMS text messaging cessation programmes. Cannabis use is not addressed through treatment but use patterns are tracked throughout.

Trial setting

This is a fully remote, decentralised clinical trial conducted in the USA. Participants are recruited nationally and complete all study procedures remotely. The trial is coordinated by the Medical University of South Carolina (MUSC). This manuscript reports protocol components according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist.

Trial registration

This study is registered on ClinicalTrials.gov (NCT06239350; registered on 2 February 2024) and is available at: https://clinicaltrials.gov/study/NCT06239350.

Primary and exploratory study outcomes

Primary study measures and outcomes include: behavioural economic assessments of demand and substitution for preferred nicotine and cannabis products (degree of substance substitutability), in addition to biochemical verification and self-reports of substance use. Specifically, the primary outcome measure includes 7-day point prevalence nicotine abstinence (PPA) at EOT (Week 12), defined as no self-reported nicotine use (cigarettes, e-cigarettes or other nicotine products) in the prior 7 days and biochemical verification via oral fluid cotinine levels below the established abstinence threshold (≤10 ng/mL). This definition of abstinence is applied consistently across all time points where abstinence is assessed (eg, weekly visits, follow-up assessments), although Week 12 serves as the primary endpoint for analysis. The secondary, exploratory outcome measure captures changes in self-reported cannabis use during the final 4 weeks of nicotine cessation treatment and is operationalised as frequency (days of use) and quantity of cannabis use (episodes of use, amount) and behavioural economic demand indices (eg, demand intensity and elasticity) over time. Secondary outcomes will assess changes in co-occurring cannabis demand and use parameters (cannabis frequency and amount) and if the time varying effects of substance substitutability modify the relationship during tobacco treatment. Other outcome measures include behavioural economic measures of nicotine–cannabis substitutability, demand and reciprocal longitudinal associations between nicotine and cannabis use through treatment (Weeks 1–12), assessed using repeated measures and cross-lagged modelling approaches (see online supplemental table S1) and/or ClinicalTrials.gov (NCT06239350) for detailed timeframe and additional information). This study protocol reflects Version 14 (February 2026) of the trial protocol.

Participants

Inclusion/exclusion criteria

Inclusion and exclusion criteria are shown in table 1, in addition to considerations or rationale for the selection of the study sample. Briefly, the sample includes those between ages 18 and 25 who are using nicotine and cannabis products regularly, while also providing biochemical verification of their use status at screening. Participants are excluded if they have unstable or significant medical or psychiatric conditions, are currently engaged in tobacco treatment or are interested in immediate cannabis cessation treatment. Given that cannabis cessation is not addressed as part of study treatment, participants seeking immediate treatment for their cannabis use at screening were not eligible to enrol.

Table 1. Full list of eligibility criteria and considerations in defining the study sample.

Eligibility criteria
Inclusion criteria	Considerations and/or rationale
Ages 18–25	Study population is focused on the priority population of 18–25 years old, given high prevalence rates of co-use in this age group.
Must use at least one nicotine product (vaping and/or cigarettes) on at least 20 out of the past 30 days for at least the past 3 months	Goal was to recruit a representative sample of young adults using nicotine regularly.
Must express interest in nicotine cessation (all products)	Nicotine cessation treatment is provided to all study participants.
Must submit a positive instant-read, oral fluid cotinine sample prior to enrolment	Confirmation of co-use status.
Must self-report cannabis use on at least 10 out of the past 30 days and/or submit a positive qualitative oral fluid cannabinoid test prior to enrolment	Confirmation of co-use status and sufficient cannabis use in the sample to evaluate study aims.
Must be able to provide informed consent	Ensuring that informed consent and participation in the study is voluntary and all procedures are understood.
Exclusion criteria
Significant or acutely unstable medical, psychiatric or substance use problem that would contraindicate research, interfere with safety, compromise data integrity or preclude consistent participation	Ensuring safety of study participants during the study and data integrity.
Pregnant (self-report) or trying to become pregnant	Ethical concerns with known cannabis use that is not being addressed among pregnant participants.
Use of medications with smoking cessation efficacy	Ensuring that participants are not engaging in treatment for nicotine use at the time of screening. Use of smoking cessation medications after enrolment is captured throughout the study.
Immediate interest in cannabis cessation treatment	Since cannabis use is not being addressed as part of study treatment, any immediate interest in cannabis treatment is deemed exclusionary and resources are provided.

Recruitment

Primary methods of recruitment include digital advertising, as well as recruitment through the MUSC Entryway programme. Entryway is a MUSC service that recruits adolescents and young adults interested in potentially participating in a substance-related research study.³⁹ Given that Entryway serves multiple trials recruiting similar populations, advertising for trials, as well as participant screening processes, are centralised. All interested study referrals complete a standardised intake protocol which has its own Institutional Review Board (IRB)-approval and consent. Eligibility criteria are concurrently assessed during intake for each of the multiple partnering studies and participants consent to share their intake data with the study in which they ultimately enrol. This centralised intake maximises the chances that a potential participant is eligible for at least one study, allows for seamless transition to a different study if a potential participant is not eligible for the originally intended study, and prevents dual enrolment across participating studies, among other potential benefits. Advertisements are developed and deployed by our study team, the Entryway team and clinical research advertising and recruitment companies and are distributed across outlets.

Procedures

Study design

Following pre-screening, completion of the Entryway intake assessment and determination of potential eligibility, participants are scheduled for a remote Screening assessment for this study (figure 1). Informed consent is obtained via a telehealth meeting with study staff before any procedures are performed (see online supplemental material for the participant informed consent form). After initial screening assessments are completed, participants are mailed instant-read cotinine and drug screen tests (for cannabis detection) to complete via video chat to confirm eligibility (pre-enrolment). On Day 0 (enrolment), participants download and are trained on how to use the MyTrials mobile application (described further below) and how to provide oral fluid samples via video capture (as part of the study intervention). The target quit date for all nicotine products starts on the morning of Day 6 and nicotine cessation treatment is provided throughout the 12-week treatment phase. Participants complete remote weekly visits during treatment. During weekly visits, staff: (1) assess changes in health or adverse events (AEs), (2) troubleshoot issues, (3) provide nicotine cessation counselling and (4) administer assessments. The EOT visit takes place on Week 12. Participants are asked to complete follow-up visits at Weeks 16 and 26.

Figure 1. Study design (timeline and overview of study procedures). Overview of the fully remote, prospective 12-week nicotine cessation trial among young adults (ages 18–25) who co-use nicotine and cannabis. Following screening and informed consent procedures, eligible participants complete pre-enrolment activities and enrolment (Day 0), including training on study procedures and use of the MyTrials mobile application. Participants initiate a nicotine quit attempt on Day 6 and receive nicotine cessation treatment throughout Weeks 1–12, consisting of digital contingency management for biochemically verified nicotine abstinence, brief weekly counselling and recommended SMS-based cessation support. Weekly study visits include assessment completion, AE monitoring, counselling and oral fluid sample collection approximately every 4 days. The primary EOT assessment occurs at Week 12, with F/U assessments completed at Weeks 16 and 26. AEs, adverse events; EOT, end of treatment; F/U, follow-up.

Nicotine cessation treatment

Digital contingency management

Digital CM is used as part of the treatment for this study, which provides financial incentives based on objective verification of nicotine abstinence and incentives based on sample submission. Digital CM is administered through MyTrials, which is a stand-alone mobile app developed to enhance remote clinical trials procedures (R41 TR004224, R42 DA062250, PI Dahne). Participants submit oral fluid cotinine samples every 4 days during treatment and receive incentives based on sample submission and biochemically verified abstinence (negative cotinine, which captures all nicotine use) according to an escalating schedule (online supplemental table S2). If a participant does not submit a sample, the sample is invalid or the sample is positive for cotinine, the incentive schedule is reset to the original amount.⁴⁰ The MyTrials app interfaces with Research Electronic Data Capture (REDCap)⁴¹ and research staff review participant videos as quickly as possible to determine abstinence status and promptly issue incentives via reloadable debit cards. A second research staff member verifies that the submitted sample is compliant, confirms the result and that payment was provided to the participant as part of intervention fidelity checks. Time from sample submission to incentive delivery is also captured. Study procedures for sample collection, submission and review are provided in online supplemental materials.

Counselling

Participants engage in psychosocial counselling starting with the pre-enrolment study visit in preparation for the quit attempt and throughout treatment at weekly study visits. Counselling is informed by a structured internal manual generated from evidence-based cognitive behavioural and motivational strategies for nicotine cessation. Core components include psychoeducation about nicotine dependence, identification of triggers, coping skills for cravings and withdrawal, cognitive restructuring, problem-solving around high-risk situations and relapse prevention. Sessions are brief (approximately 5–10 min), delivered weekly by trained bachelor’s-level research staff and supervised by a licensed clinician to ensure fidelity. Session content is standardised across participants while allowing flexibility to tailor examples to individual nicotine use patterns (eg, vaping, smoking or dual use). After study enrolment commenced, session text, content and prompts were added to the study REDCap database at each weekly event to ensure fidelity and consistency. Time fields were also added so that total time of counselling is collected at each weekly visit.

Discontinuation of the intervention

Participants may discontinue study participation at any time. If a participant no longer wishes to engage in nicotine cessation, they are encouraged to remain in the study and continue completing study assessments. The intervention is still provided, and nicotine cessation interest is revisited and encouraged throughout treatment. However, the participant may withdraw from the study at any point. Termination of study participants (determined by the Principal Investigator (PI])) is rarely implemented, but participants can be terminated from the study if safety concerns arise, including the emergence of significant medical or psychiatric conditions that may interfere with participation.

Adherence, retention and follow-up

Adherence to the intervention is monitored, encouraged and captured in several ways. Participants receive regular prompts through the MyTrials mobile application to complete required submissions, and weekly study visits provide additional support through counselling to improve adherence. Adherence is monitored through multiple methods, including video-verified oral fluid cotinine sample submissions through the MyTrials platform, completion of daily diary assessments and attendance at scheduled study visits with research staff. Staff review submitted materials to verify compliance with study procedures and provide follow-up, as needed. Participants who are not adherent to study procedures are encouraged to continue completing study assessments and retraining is provided. Outcome data (ie, self-reported daily diary data, biochemical verification) continue to be collected, when available. Analyses will include all enrolled participants regardless of adherence to the intervention, when possible.

Nicotine cessation resources and concomitant care

Depending on the participant’s use of cigarettes, e-cigarettes or dual use, evidence-based nicotine cessation resources are recommended to augment CM and counselling. This is Quitting (SMS text-based programme)^{8 42} is recommended, in addition to SmokefreeTXT (SMS-text based programme).⁴³ Self-reported use of these resources is captured throughout the study and at follow-up visits. Cannabis use is not treated as part of the study and is monitored throughout participation. Participants are excluded at screening if they are currently using medications with established smoking cessation efficacy or are actively seeking treatment for cannabis use. Use of smoking cessation medications initiated after enrolment is not prohibited but is assessed throughout the study. Specifically, use of non-study nicotine treatment initiated after enrolment (eg, pharmacotherapy or external programmes) is assessed at the EOT study visit and at follow-up. Sensitivity analyses may be conducted to evaluate whether results differ when excluding participants who initiate external cessation treatments during the study.

Instruments and measures. A complete listing of instruments/measures is shown in online supplemental table S3

Screening/baseline assessments

Demographics, substance use and other clinical characteristics are collected at screening. Current substance use (vaping, smoking, other nicotine, alcohol, cannabis and other substance use) is collected via 30-day timeline follow-back (TLFB).⁴⁴ To capture self-reported cannabis use, we determine an individual-specific cannabis unit, which has been used successfully in ongoing and recently completed studies⁴⁵ (eg, R37 CA237245; R01 DA042114; R01 DA054617). For plant material-based methods, participants are asked to estimate the average amount of cannabis (grams) used for each modality (eg, joints). That amount then serves as the individual’s standard unit. For non-plant-based methods (oils, concentrates, etc), we assess times used or hits taken (waxes, dabs), milligrams ingested (edibles) and millilitres used (oils, tinctures). We also capture hemp-derived alternative cannabinoid products (eg, Δ⁸-tetrahydrocannabinol [Δ⁸-THC], Δ¹⁰-THC).

Biochemical verification

Nicotine use is captured with qualitative and quantitative oral fluid cotinine tests.⁴⁶ Qualitative cotinine is collected via instant-read oral fluid tests (cut-off of 30 ng/mL) to determine eligibility and abstinence during treatment. Qualitative, instant-read urine drug screens are used to assess for THC to determine eligibility (cut-off for THC is 50 ng/mL), with results shown to research staff virtually to confirm eligibility. Oral fluid samples are collected at key time points (see online supplemental table S3) and shipped to Forensic Fluids Laboratories, where they are tested for: semiquantitative cotinine and anabasine (alkaloid that distinguishes combustible tobacco), Δ⁹-THC, cannabidiol, Δ⁸-THC and other cannabinoids.

Cannabis and nicotine demand

Cannabis and nicotine purchase tasks are administered to assess demand, as well as bidirectional substance substitution, during the screening and enrolment sessions, Week 4 and 8, EOT and the Week 26 follow-up.⁴⁷,⁵⁰ Hypothetical purchase tasks (HPT) assess how many of a given product or product combination(s) participants would purchase at specified prices, yielding data that: (1) are clinically relevant indicators of valuation processes, (2) correlate with measures of quantity and use disorder symptom severity and (3) are reactive to experimental manipulations that alter use patterns.^{35 51 52} HPT assessments are administered through the Healthful Experimental Store.^{47 53} The instructions, price range and description of consumption units are based on procedures established for validated and clinically relevant cigarette and marijuana purchase tasks.^{49 50} Nicotine products include cigarettes (one pack of cigarettes) and nicotine vapes (standardised as vapes that include a 1 mL cartridge or around 200 puffs). Cannabis products include combustible cannabis (flower, 1 g), concentrates (1 g) and edibles (preferred type and THC content). This allows us to calculate demand for individual nicotine products, as well as demand for all nicotine products collapsed together, based on total nicotine content.⁴⁷ Similar strategies are employed for cannabis to capture demand for individual products, as well as demand for cannabis products as a whole based on total THC content. Cross-commodity purchasing tasks assess hypothetical demand for nicotine and cannabis under varying price conditions, allowing assessment of the degree of substitution (eg, increased cannabis demand when nicotine becomes more expensive), complementarity (increased demand for one commodity when the other is less expensive) or independence between the two substances.

Mobile daily diaries

Using REDCap⁴¹/Twilio integration, a survey link is sent via text message to participants every morning to assess their substance use (eg, vaping, cigarettes, other nicotine, cannabis) the previous day, which has been used successfully in other studies.^{54 55} Missing data are collected at the next study visit using TLFB procedures.

Adverse events

AEs and serious AEs (SAEs) are systemically collected at weekly study visits via participant self-report. Participants are asked to report any hospitalisations or visits to the emergency department since the last visit and any other significant events that affected their daily functioning. Research staff collect additional information about reported events. Study medical clinicians review all events, determine relatedness to the study and intervention and discuss with the participant, as needed, to gain more information. The potential risks associated with this study are expected to be minimal and include nicotine withdrawal, loss of confidentiality and discomfort regarding interview questions.

Participant and public involvement

For this particular study, participants or the public were not involved in the design, conduct or reporting of this research. For past co-use studies, we have obtained feedback from community members on recruitment materials, study design features, etc, to improve enrolment and adherence. We may involve community members or outreach in the dissemination of study findings.

Participant compensation

Participants are compensated for study visits on the following schedule: US$30 for screening, US$20 for training, US$40 each for enrolment (Day 0) and the Week 12 and 26 follow-up visit (US$120 total), US$15 for the rest of the weekly study visits except for Week 4 and 8 (US$30 each) and US$10 for the Week 16 follow-up visit (US$375 total for study visits). To enhance completion of the Week 12 EOT visit, a US$50 bonus was implemented after study launch, which includes an Amazon gift card on completion of all study procedures at the EOT study visit. Participants are compensated for completing daily dairies (US$2 per diary completed, paid weekly). Incentives (US$445 possible) are delivered starting on Day eight through Day 84 on an escalating schedule (online supplemental table S2). After study enrolment launch, we modified the incentive schedule for MyTrials. Participants now receive US$4 for sample submission, while incentives for abstinence remain the same. Participants may now earn US$465 for MyTrials submissions and nicotine abstinence. To encourage the shipment of oral fluid specimens to the external laboratory, part of study visit compensation was made contingent on evidence of specimen shipment, either through a photo confirming shipment or obtaining results from the laboratory (US$10 per sample confirmation). All compensation is provided through reloadable debit cards. Participants may earn a total of US$1030 during the study.

Statistical analyses

Power and sample size

Sample size estimation and power was based on Aim 1, which will evaluate the association of behavioural economically derived measures on substance substitutability on Week 12 EOT nicotine abstinence. Our hypothesis is that participants with greater substitutability patterns will have lower rates of 7-day PPA from nicotine at EOT compared with patterns indicating a complementary relationship. To inform sample size estimation, related research studies from the co-use treatment literature were explored to estimate EOT nicotine cessation rates. First, in a cross-sectional survey of national household adult (ages 18+) data,⁵⁶ tobacco cessation was reported among 21.8% in those reporting co-occurring nicotine and cannabis use. Second, in a youth varenicline trial,¹⁶ at EOT, differences in tobacco PPA between those co-using cannabis and those only using tobacco was present (26.0% vs 12.2%; Δ=13.8%; Risk Ratio (RR)=2.2 (1.1–4.5); p=0.025).¹⁷ Third, in a remote CM study for smoking cessation among adults,⁵⁷ CM produced nearly 40% abstinence at 4 weeks (primary outcome) and 29% abstinence at the 12-week follow-up visit. The current study protocol assumes that nicotine treatment among young adults who are co-using nicotine and cannabis will achieve a similar 15–20% abstinence rate at EOT. To detect a clinically relevant 7% increase in EOT abstinence^{8 58 59} associated with a SD increase in baseline substitutability, a sample size of n=280 participants will be necessary to achieve 80% power with a 5% type 1 error rate. Assuming an attrition rate of 20% at EOT, a total inflated sample size of N=350 will be necessary to maintain adequate statistical power. Aim 2 (related to cannabis demand and use during nicotine treatment) will determine if nicotine abstinence, reduction or withdrawal observed during treatment (1) is associated with changes in cannabis demand and use (frequency, quantity) and (2) if substitutability modifies this relationship. Our hypotheses are that the relationship between nicotine abstinence, reduction or increased nicotine withdrawal will be significantly associated with increased cannabis demand and use and that the relationship will be moderated by time varying changes in substitutability. Although not powered to detect a specific interaction effect, a sample size of N=350 enrolled participants (n=280 treatment completers) will have 80% power with a type 1 error rate to find substitutability (substitutes vs compliments) moderated effect size of the association of nicotine abstinence, reduction and withdrawal on cannabis outcomes of standardised β=0.34. Aim 3 is exploratory and therefore, we are not powered to find reciprocal relationships between nicotine and cannabis use during treatment.

Statistical analyses

Behavioural economic demand analyses

Behavioural economic analyses involve the use of demand parameters (eg, demand elasticity, demand intensity, substitutability) in predictive statistical models. These models require demand parameters at the individual participant level. Demand data requires non-linear modelling to accurately describe the relationship between cost and consumption. Statistical techniques specific to behavioural economics and substitution previously described³⁵⁶⁰,⁶³ and successfully used on projects with similarly complex statistical analysis plans^{47 53 64 65} will be implemented. Demand data will be fit to a modification⁴⁸ of a model proposed by Hursh and Silberberg⁶⁶ to quantify the relationship between the price of a commodity and consumption of that commodity:

$${\displaystyle {\displaystyle Q=Q_0\ast {10}^{k(e^{-\alpha Q_{0x^c}}-1)}}}$$

where Q is consumption of the commodity, C is the price, Q₀ is the derived initial consumption without cost constraints (demand intensity), k is the span of the function in logarithmic units and α is the demand elasticity. Q and C are determined by the data and k is set to a constant determined empirically by the actual data, leaving only Q₀ (demand intensity) and α (demand elasticity) as free parameters to be fitted. Analyses will be conducted using non-linear mixed effects modelling.⁶³ Mixed effects modelling is preferred over traditional analysis approaches because these models allow for clustering specification of individuals, time and experimental conditions while also allowing for inclusion of covariates and model comparisons.⁶⁴ Further, mixed effects models can be used to estimate individual-subject random effects, thereby providing a framework to determine demand parameters for each individual in an experiment in a common model with common assumptions. Consumption data for both nicotine and cannabis will be analysed for all participants in a single mixed effects model with participant entered as a random effect. These participant-level random effects for demand elasticity (α) and demand intensity (Q₀) will then be extracted for subsequent analyses.

Substitution analyses

This project involves characterising the degree to which cannabis is a substitute or complement for nicotine and vice versa (degree of substance substitutability). To address the substitution goals, a modification analogous will be used⁴⁸ of an equation proposed by Hursh and Roma,⁶⁷ which relates consumption of a fixed-price drug as the price of an alternate substance changes:

$${\displaystyle {\displaystyle Q=Q_{\text{alone}}\times {10}^{I\times e^{-\beta C}}}}$$

In this equation, Q_alone represents the maximum consumption of the fixed-price commodity when the primary commodity has a price high enough to drive consumption to zero, ß is the sensitivity of fixed-price commodity consumption to the price of the primary commodity, I is an interaction parameter that indicates whether the fixed-price commodity is a substitute (positive I value), complement (negative I value) or independent (I value near zero), and Q and C represent consumption of the fixed-price commodity and cost of the primary commodity, respectively, and are known from data. This equation has three parameters (Q_alone, I and ß) that will be estimated from experimental data using non-linear regression, although the I parameter is the focus of the proposed aims. Analyses will proceed similarly as the demand analyses above with the substitution equation being fit using mixed effects modelling. Values will then be further analysed to satisfy the goals of each aim.

In summary, behavioural economic indices are operationalised at the individual participant level. Demand intensity (Q₀) reflects estimated consumption at zero (or no) cost, while demand elasticity (α) reflects sensitivity of consumption to increasing price. Substitutability is indexed by the interaction parameter (I) derived from cross-commodity purchase tasks, with positive values indicating substitution, negative values indicating complementarity and values near zero indicating independence between nicotine and cannabis. Values greater than zero will be interpreted as substitution, values less than zero as complementarity and values not statistically different from zero as independence. These parameters are estimated using non-linear mixed-effects models and participant-level random effects are estimated for each variable for use in subsequent regression and longitudinal analyses. Specifically, in primary analyses, substitutability will be treated as a continuous variable and secondary/exploratory analyses will examine categorical groups (substitution, complementarity, independence). As substitutability may vary over time, time-varying estimates will be used in longitudinal models. Additional sensitivity analyses will examine baseline substitutability and stability of classification schemes over time to assess whether shifts are associated with changes in nicotine and/or cannabis use.

Baseline characteristics

Demographic, clinical and baseline substance use characteristics will be summarised for the entire cohort. Purposeful selection of covariates for each model will be implemented as described by Bursac et al.⁶⁸ Possible covariates include age, sex, race, ethnicity, nicotine and cannabis use characteristics, nicotine dependence and the marijuana craving questionnaire total and subscale scores. Candidate variables will be first screened using univariable analyses with a relaxed inclusion criterion (p<0.2), then entered into a multivariable model and sequentially removed if not significant (p>0.05), while assessing possible confounding on primary predictor coefficients (∼15% or more change in parameter estimate). All models will include a participant-level cannabis regulatory environment covariate (legality status in their state of residence).

Primary and secondary outcome analyses

Primary and secondary/exploratory outcomes are noted above in Section 2.1.2.

Aim 1

To assess the association between the behavioural economic measure of baseline substance substitutability (I) of nicotine and cannabis and the log-odds of nicotine abstinence at EOT (biologically-confirmed 7-day PPA from nicotine), logistic regression models will be developed. Nicotine abstinence criteria for quantitative oral fluid cotinine will be set at 10 ng/mL.⁴⁶ Baseline levels of cannabis use, nicotine dependence and nicotine use type (smoking, vaping or both) will be included as primary covariates.⁶⁸ Demographic and clinical correlates significantly associated with abstinence will be included in all adjusted models. Continuous variables will be standardised prior to analysis and will be assessed for linearity with the logit through the Box-Tidwell test and visual examination of data plot.

Aim 2

To examine whether nicotine abstinence, reduction or withdrawal during treatment is associated with co-occurring cannabis demand (Q₀, α) and cannabis use (frequency, quantity) and if time varying effects of substitutability modify this relationship, generalised linear mixed effects models will be developed. Specifically, Aim 2a will assess the relationship of nicotine abstinence, reduction or withdrawal separately on cannabis demand and use. Aim 2b will then assess if substitutability, as determined for Aim 1, moderates these associations. Nicotine and cannabis use will be collected through daily diaries and tabulated weekly. Substitutability and demand will be assessed at baseline, and Weeks 4, 8 and 12 (EOT). Continuous use and demand values will be standardised and centred within each participant such that negative values will represent lower than average and positive values will represent higher than average values for each participant. Additionally, grand mean centred values will be included as a time invariant value in all models. Models will assess the association of nicotine use on cannabis use and demand, visit and the moderating effect of substitutability (through an interaction term). When the interaction is significant, the model will be stratified by whether substances are substitutes or complements. Restricted maximum likelihood (ML) methods will be used to estimate the fixed effects and variance components, and model-based effect estimates will be used to construct group level tests over time when appropriate. Many covariance structures will be considered, and Akaike’s information criteria will be used to select the optimal covariance pattern for the within-subject data. Our hypotheses are that (1) nicotine abstinence, reduction and withdrawal will all be associated with increased cannabis demand and use and (2) those associations will be moderated by substitutability.

Aim 3

To investigate the reciprocal prospective relationship between nicotine and cannabis use patterns during treatment, structural equation models using an ML cross-lagged panel design will be developed.⁶⁹ Nicotine and cannabis use (frequency, quantity) will be collected at baseline and weekly using tabulated daily diary data. Values will be standardised and centred within each participant such that negative values will represent lower than average and positive values will represent higher than average values. Additionally, grand mean centred values will be included as a time invariant variable. Autocorrelations and cross-lagged correlations will be modelled, allowing concurrently measured data to have a synchronous relationship (comovement). In addition to the fully reciprocal model, autocorrelation only and single variable lagged models will be constructed and compared with the initial model using various fit statistics; specifically, the Comparative Fit Index (CFI),⁷⁰ Tucker-Lewis Index (TLI)⁷¹ as well as the root mean squared error of approximation (RMSEA).⁷² RMSEA values less than 0.06 and CFI/TLI values greater than 0.90 indicated adequate model fit. We hypothesise that for those participants showing high substitutability, weekly nicotine and cannabis use will diverge, while those with a complementary relationship will show similar weekly use patterns. We are not powered to detect an effect for exploratory Aim 3.

Sex as a biological variable

Sex as a biological variable will be assessed as a predictor in all models. When significant, sex differences and stratified parameter estimates with associated CIs will be reported. Further, we will examine sex differences in baseline substitutability and demand.

Missing data for analyses and multiplicity

The study team has procedures in place to minimise missing data and attrition (eg, reminders, detailed locator information); however, these methods do not ensure that all data will be collected. Baseline demographic and clinical characteristics will be collected on all enrolled participants. We will make every effort to continue assessments for the entire study, even among those who stop participating. Even with these methods in place, statistical plans assume an attrition rate of 20% at EOT for study outcome measures.

Appropriate analysis methods will be employed to accommodate missing data. Methods of ML yield valid inferences assuming ignorable attrition (ie, attrition is accounted for by covariates or the dependent variable measured prior to dropout). Analyses will be conducted under a missing at random assumption using ML estimation in mixed-effects and longitudinal models. To evaluate the robustness of this assumption, (1) the proportion of missing outcome data will be compared across cohorts and (2) sensitivity analyses will be conducted, including models incorporating predictors of missingness. For binary abstinence outcomes, participants with missing outcome data will be classified as (1) non-abstinent as well as (2) abstinent in primary analyses, with sensitivity analyses evaluating these alternative assumptions. If evidence is present that the assumptions are violated, methods of multiple imputation will be employed and estimates compared with ML results.

Given the multiaim approach of the study, primary inference for study aims will use a critical alpha of 0.05 for all models. Secondary and exploratory analyses will be interpreted with consideration of multiple comparisons and adjustments will be applied using the Benjamini-Hochberg false discovery rate (5%).⁷³

Data collection

Data management

Data are collected and managed by trained study staff using secure, institutionally approved platforms, including REDCap. Personal information is collected during screening, enrolment and study procedures. Participant data are identified using unique study IDs, and identifying information is stored separately from study data. Data are entered directly by participants (eg, daily diaries) or by study staff and are stored on secure, password-protected servers maintained by MUSC. REDCap includes built-in data validation procedures (eg, required fields, logic checks) to minimise entry errors, and audit trails are used to track data entry and modification. Access to study data is restricted to authorised personnel. Following study completion, de-identified data may be shared with other researchers in accordance with institutional policies and data use agreements. Identifiable information will not be shared outside the study team and will be maintained securely in accordance with institutional and regulatory requirements.

Trial progress and monitoring

MUSC IRB approval was received on 15 September 2023, as an expedited study protocol (reference ID: Pro00131124) and all annual continuing reviews have been approved by the MUSC IRB. Study enrolment began in May 2024, with the first participant being consented on 21 May 2024 and the first participant being enrolled in the study on 4 June 2024. As of April 2026, we have enrolled 137 of the proposed 350 participants into the study (39%). Enrolment is expected to conclude in September 2027.

Trial conduct is monitored on an ongoing basis by the study team and PI. Study staff conduct weekly monitoring through participant visits, which includes assessment of AEs, protocol adherence and data completeness. The PI performs regular reviews of study data and reported AEs to ensure participant safety and data integrity. In addition, the Data and Safety Monitoring Board (DSMB) reviews study progress, safety data and overall trial conduct at regular intervals (at least annually). These procedures ensure continuous oversight of trial conduct, data quality and participant safety throughout the study.

Trial oversight and management

This trial is coordinated by MUSC. The PI is responsible for overall study design, conduct and oversight. The study team (ie, coinvestigators and research staff) supports day-to-day trial operations, participant recruitment, data collection and cessation intervention delivery. Data management is conducted by the study team using secure, institutionally approved platforms (eg, REDCap). Oversight of study conduct, data integrity and participant safety is managed by the PI and study team, with additional institutional oversight provided by the MUSC IRB and an independent, internal DSMB.

Data and safety monitoring

The DSMB for this project provides oversight of study progress, data quality and participant safety on an annual basis, or more frequently, as needed. The DSMB is composed of individuals with relevant expertise in clinical trials, substance use research and biostatistics who are not otherwise involved in the conduct of the study. The DSMB provides recommendations to the PI following review of safety data and study progress and members report no conflicts of interest.

Interim analyses and stopping guidelines

No formal interim analyses for efficacy are planned for this study. Safety data are reviewed on an ongoing basis by the PI and annually by the DSMB, which may recommend modification or early termination of the study in the event of unexpected safety concerns or other issues affecting study feasibility or integrity. The PI, in consultation with the DSMB and institutional IRB, holds final responsibility for decisions regarding trial continuation or termination.

Design decisions, challenges, considerations and modifications

Inclusion of multiple forms of nicotine use

In determining the most representative sample for this study, the use of multiple nicotine products was considered and determined to be inclusionary. Polyproduct nicotine use is common among young adults^{74 75} and is largely composed of vaping nicotine, smoking cigarettes or both. Limiting the study sample to those who only vape or only smoke cigarettes would jeopardise the impact of findings and reduce generalisability. Behavioural economics analyses allow for the assessment of dual substance use patterns at the individual level, regardless of nicotine product used, through extracted measures of demand and substitutability. That allows for testing stated hypotheses and exploratory analyses to compare substitutability across nicotine product types. Failure to capture real-world patterns of nicotine use among young adults and evaluate the underlying relationship with cannabis was determined to severely affect the impact of findings and our ability to inform inclusive treatment recommendations.

Quantifying cannabis use and defining co-use status

There are no standard definitions of cannabis-nicotine co-use. To allow for the most representative sample of young adults who are co-using, participants must meet a minimum threshold of cannabis use to be included; however, this threshold allows for variability in co-use patterns. Previous studies have categorised cannabis use severity based on national household survey recommendations,^{11 13} with 20 days of use in the past 30 indicating regular use, while intermittent use is defined as some past month use, but less than 20 days out of the past 30. Participants in this study must have used cannabis (in any form) on at least 10 days out of the past 30 and/or must test positive on an instant-read cannabinoid assay (pre-enrolment visit). Self-reported cannabis use, in conjunction with biochemical verification, will contribute to appropriate enrolment and severity categorisation.

Remote trial methodology

Remote trial methodology comes with increased accessibility and reach but can also lead to potentially greater rates of attrition and lower rapport with study staff. Further, participants are asked to complete tasks outside of study visits, such as oral fluid collection, recording and uploading videos to a mobile app, shipping oral fluid samples in a timely manner after the study visit, etc. Therefore, compliance with study procedures may be more difficult than in in-person studies and we may have increased rates of missing data. An additional consideration of the fully remote design is the potential for selection bias related to digital access, technological literacy and participant engagement. Individuals with limited access to smartphones, internet connectivity, private environments for study participation or lower comfort with digital platforms may be underrepresented, which could affect the generalisability of findings. However, remote trial methods may reduce other sources of participation bias by minimising geographical, transportation, scheduling and time-related barriers that can limit enrolment in traditional site-based studies and aid in the recruitment of representative samples.⁷⁶ This approach may increase access to individuals who would otherwise be unable to participate in in-person research and supports recruitment of a geographically diverse sample of young adults across the USA. Therefore, while remote methods introduce unique considerations for representativeness and engagement, they may also improve accessibility and reduce barriers to participation for some populations.

Addressing cannabis use

Cannabis use is not the focus of study treatment, though to ensure the safety of participants, all are given information regarding cannabis at screening and participants are excluded if they are seeking immediate treatment for cannabis. If the participant expresses a desire to reduce/quit using cannabis during the study, treatment resources are provided (hand-outs, websites to access treatment, etc) but kept minimal to avoid researcher-imposed impact on cannabis. The only form of cannabis use that is addressed as part of the intervention is specific to blunt use. Blunts are hollowed out cigars/cigarillos filled with loose-leaf cannabis, which expose participants to tobacco via the wrapper and may interfere with the ability to detect nicotine abstinence (via elevated cotinine levels). The inclusion of blunt use among participants in this sample was deemed essential given; (1) that racial/ethnic differences in blunt use have been demonstrated,⁷⁷ (2) among young adults who co-use, blunts were rated as the second most popular tobacco product (behind cigarettes),⁷⁸ and (3) excluding young adults who use blunts would be detrimental to generalisability. Blunt use, as it affects cotinine levels and detection of abstinence, is discussed with participants as part of nicotine cessation counselling.

Participants are not seeking treatment for their cannabis use at baseline and to maintain the integrity of the study design and aims, we do not intervene on cannabis use through treatment. Participants who are enrolled are not discouraged from quitting/reducing cannabis use (and resources are provided), but cannabis use is not addressed through the study. The goals of the study are to understand individual differences in cannabis co-use on nicotine outcomes to inform tailored treatment and therefore, enrolling those not seeking treatment for cannabis and without any desire or plans to quit was deemed the most appropriate design decision for the study aims and ethically justified.

Protocol modifications implemented after enrolment launch

Several protocol amendments have been submitted since the launch of study enrolment. All protocol modifications (eg, changes to eligibility criteria, outcomes or analyses) are submitted and approved by the institutional IRB prior to implementation. Protocol modifications are updated in ClinicalTrials.gov (NCT06239350) and communicated to study personnel. Retraining is provided, if necessary. When relevant, participants are informed of changes to the protocol that may affect their participation and reconsent is obtained if required.

Several notable protocol modifications have been made to the study to enhance adherence and minimise missing data. Incentives for oral fluid sample submissions through MyTrials were increased from US$3 to US$4. Given minimum amounts of participant compensation that can be loaded onto reloaded debit cards used in this study, we increased sample submission compensation to allow for immediate loading of the incentive onto the debit card, without delay, to more effectively reinforce submission. An EOT study visit bonus was implemented (US$50 Amazon gift card) to encourage attendance at that study visit and completion of all study procedures. Additionally, the total number of quantitative oral fluid assays being collected has been reduced in recent amendments to reduce burden on participants and study staff, while still maintaining integrity of the data collection.

Limitations

Additional limitations should be considered. First, the sample may reflect a degree of selection bias toward young adults who are motivated to quit nicotine, which may limit generalisability to individuals with lower readiness to change. Second, as with many longitudinal trials, differential attrition may occur and participants who discontinue participation may differ systematically from those who remain engaged, which would potentially introduce bias despite the use of sensitivity analyses. Additionally, although repeated longitudinal assessments allow examination of temporal ordering and prospective associations between nicotine and cannabis use, the study is observational with respect to cannabis exposure and cannot establish causal relationships regarding the effects of cannabis use patterns on nicotine cessation outcomes. Third, measurement limitations should be acknowledged. Self-reported cannabis use may be subject to misclassification, particularly for products such as blunts that combine cannabis and tobacco. Additionally, while cotinine-based biochemical verification is widely used, it may not fully distinguish nicotine abstinence from low-level or intermittent exposure (eg, secondhand smoke or nicotine replacement therapy use), which could result in misclassification of abstinence status. Finally, and importantly, study results will not be able to make causal inferences on the role of cannabis co-use on nicotine cessation. Cannabis is not being controlled experimentally and there are several other variables that may be affecting the cannabis use and nicotine cessation association, such as co-occurring psychiatric disorders, severity of use, and motivation to quit nicotine. We acknowledge these limitations to study results. However, results will uncover unique characteristics and aspects of the cannabis–nicotine co-use relationship that may be associated with treatment outcomes, and therefore, lead to treatment tailoring and recommendations for young adults who co-use.

Ethics and dissemination

The PI, the entire research team and the university in which this research is being conducted, are committed to the efficient dissemination of study findings. All members of the research team agree to adhere to the National Institutes of Health (NIH) ‘Principles and Guidelines for Recipients of NIH Research Grants and Contracts on Obtaining and Disseminating Biomedical Research Programs.’ A final de-identified data set will be made available for retrieval and analysis of secondary outcomes, on request to the PI and/or corresponding author. All manuscripts published from these data will be submitted to PubMed Central.

This study has received IRB approval from MUSC and all participants provide informed consent prior to study procedures. The PI has overall responsibility for safety and data monitoring on a day-to-day basis for this trial. The PI provides weekly checks of reported AEs to determine if particular categories are being endorsed more frequently than anticipated. This study includes a behavioural intervention given to all study participants, with no medication provided and as such, AEs and SAEs are expected to be rare or unrelated. Participant consent permits the use of de-identified data and biological specimens for future research studies. Identifiable information will be removed prior to any such use or sharing and additional informed consent will not be required for secondary analyses conducted in accordance with institutional and regulatory guidelines. Given the behavioural nature of the intervention, risks are expected to be minimal and largely include the potential for loss of confidentiality and nicotine withdrawal. Participants who experience AEs or express a need for additional care during the study are provided with appropriate referrals for medical, psychological or substance use treatment services as needed.