Determining the impact of cannabis use on tobacco cessation: Results from a prospective non‐randomized tobacco treatment trial

Abstract Background and aims Tobacco‐cannabis co‐use is associated with increased psychosocial and health harms; however, the treatment literature assessing the impact of co‐use has been mixed and suffers from critical limitations. To date, no prospective studies have evaluated the degree to which cannabis co‐use affects tobacco cessation, nor are there treatment recommendations for those who co‐use. The goal of this study was to evaluate and quantify the impact of cannabis co‐use on tobacco cessation through a 12‐week prospective tobacco cessation trial. Design This was a multi‐site, non‐randomized trial. Inverse probability weighted modified Poisson regression models were used to assess the probability of tobacco abstinence between tobacco‐only and tobacco‐cannabis co‐use cohorts. Setting Enrollment occurred at three sites across South Carolina, USA: The Medical University of South Carolina (MUSC) Charleston, Behavioral Health Services of Pickens County and MUSC Florence. Participants Enrolled participants were adults ( n = 181; ages 18–40) who (1) smoked tobacco cigarettes and used cannabis regularly (co‐use cohort; oversampled) or (2) smoked tobacco cigarettes daily (tobacco‐only cohort). Across 3 sites, 181 adults (110 co‐using cannabis; 71 tobacco‐only) were enrolled (49% female; 19% Black/African American; 6% Hispanic/LatinX). Interventions All participants received 12 weeks of first‐line tobacco treatment (varenicline, incentives for tobacco abstinence, counseling), while cannabis use was not addressed as part of treatment. Measurements The primary outcome was biochemically confirmed 7‐day point prevalence tobacco abstinence (PPA) at end of treatment (EOT; Week 12 post‐enrollment). Findings Average baseline reported cigarettes per day were not different between tobacco‐only or cannabis co‐use cohorts [14.9, (standard deviation, SD = 7.2) versus 13.4 (SD = 7.8); P = 0.25]. Among the co‐use cohort, days of cannabis use in the past month at baseline was 23.7 (SD = 9.5). In imputed analyses, EOT 7‐day PPA tobacco abstinence was achieved in 59% of tobacco‐only participants and 33% of the cannabis co‐use cohort [Imputed data: relative risk (RR) = 1.63, 95% confidence interval (CI) = 1.14–2.33]. Neither baseline daily cannabis use [adjusted relative risk (RR adj ) = 1.05, 95% CI = 0.63–1.77] nor within‐treatment cannabis use frequency (RR adj = 0.51, 95% CI = 0.25, 1.06) was associated with EOT PPA for tobacco in the cannabis co‐use cohort. Conclusions This is the first prospective study designed to compare tobacco cessation outcomes by cannabis co‐use status. Regular cannabis co‐use had a negative impact on tobacco abstinence among younger to middle‐aged adults. Addressing cannabis use should be incorporated into standard tobacco treatment to improve cessation outcomes.