Cannabis is the most commonly used substance among individuals who drink alcohol in the United States (Subbaraman and Kerr, 2015). Co-use prevalence has been increasing among adults, concomitant with the widespread legalization of recreational cannabis (Gonçalves et al., 2023). Existing studies provide conflicting results regarding the influence of cannabis on alcohol consumption; some data support complementarity between alcohol and cannabis, such that cannabis use is associated with increased alcohol consumption, while others support substitution effects, such that cannabis use is associated with decreased alcohol consumption (Gunn et al., 2022; Reiman, 2009; Risso et al., 2020; Subbaraman, 2014). Investigating individual difference factors and/or contextual variables that may influence the cannabis-alcohol relationship is a growing research priority with potentially important implications for harm reduction.

Several studies have found associations between co-use and increased alcohol consumption, particularly for simultaneous use (i.e., use at the same time such that intoxicating effects overlap) (Boyle et al., 2024, Gette et al., 2024, Hatch et al., 2023, Jackson et al., 2020, Lee et al., 2020, Lee et al., 2022, Looby et al., 2021, Midanik et al., 2007, Subbaraman and Kerr, 2015), however, results are nuanced and there is not consistency across the literature. For example, cannabis use has been associated with reduced abstinence from alcohol in AUD treatment-engaged populations, with some variability according to cannabis use frequency (Metrik et al., 2018, Mojarrad et al., 2014, Subbaraman et al., 2017, Subbaraman et al., 2024, Subbaraman et al., 2025). In contrast, reductions in alcohol consumption have been observed amongst individuals who use medical cannabis (Coelho et al., 2023, Gunn et al., 2019, Karoly et al., 2020, Subbaraman and Kerr, 2018), suggesting certain populations may be more likely to engage in substitution.

The majority of evidence for substitution comes from epidemiological studies of cannabis legalization (Guttmannova et al., 2016, Risso et al., 2020), however supporting evidence comes from recent self-report and observational studies (Karoly et al., 2021, Karoly et al., 2023, Pauly et al., 2021, Wilkins et al., 2024). In a study of treatment-engaged individuals who engage in heavy-drinking, Karoly et al. (2021) found that individuals drank significantly less on cannabis use days compared to non-use days, regardless of cannabis use frequency (Karoly et al., 2021). Another study found that each additional day of concurrent or simultaneous use was associated with reduced cannabis and alcohol consumption in a typical week amongst undergraduates (Karoly et al., 2023). Furthermore, in a study of non-treatment seeking youth, cannabis abstinence was associated with increased alcohol use, which returned to baseline following the resumption of cannabis use (Schuster et al., 2021). These findings are consistent with self-reports of cannabis as a harm reduction substitute for alcohol in both community and treatment-engaged samples (Pauly et al., 2021, Wilkins et al., 2024).

While human laboratory studies are comparatively limited, Mello and colleagues instructed participants to complete a basic operant task wherein they could earn cannabis or money after 30 min of work and alcohol after 15 min, and found that individuals self-administered less alcohol when cannabis was concurrently available (Mello et al., 1978). Later, Ballard and DeWitt found that co-administration of low doses of THC and alcohol reduced alcohol craving (Ballard and de Wit, 2011). An unpublished laboratory study also observed that cannabis use was associated with acute reductions in alcohol intake and some measures of craving in a within-subjects design with individuals who engage in heavy drinking (Metrik et al., unpublished). Other observational studies have reported similar reductions in alcohol consumption when cannabis was used before alcohol (Gunn et al., 2021; Karoly et al., 2023, Karoly et al., 2024; Rootman, 2020). The effects of cannabis on alcohol craving and consumption during simultaneous use may be mediated in part by subjective effects (i.e., relaxation versus stimulation), as simultaneous use predicted discontinuation of drinking via increased relaxation and decreased alcohol craving at the event-level in an ecological momentary assessment study (Waddell et al., 2024). Together, these findings suggest the potential for cannabis to act as a substitute for alcohol. Further research is needed to elucidate variables that may predict such substitution behavior.

The present study addresses the first aim of a larger trial (NCT04998006), leveraging data from a within-subjects human laboratory drug administration paradigm to explore the effects of legal-market flower cannabis use on alcohol self-administration and craving in a sample of individuals who regularly use cannabis and engage in heavy drinking. Consistent with substitution, we hypothesized that participants would self-administer fewer drinks and report less alcohol craving during a lab session in which they self-administered both alcohol and cannabis compared to a lab session in which they self-administered only alcohol.