PURPOSE: The purpose of this review is to discuss the literature regarding the concurrent use (co-use) of alcohol and cannabis and competing hypotheses as to whether cannabis acts as a substitute for (i.e., replacing the effects of alcohol, resulting in decreased use) or a complement to (i.e., used to enhance the effects of alcohol, resulting in increased use) alcohol. The impact of cannabis use on alcohol-related outcomes has received increased attention in the wake of ongoing legalization of cannabis for both medical and recreational purposes. Evidence for both hypotheses exists in the literature across a broad range of data collection methods and samples and is carefully reviewed here. In addition, various mechanisms by which cannabis may act as an alcohol substitute or complement are explored in depth with the goal of better understanding equivocal findings.

SEARCH METHODS: This review includes articles that were identified from a search for studies on alcohol and cannabis co-use, with a specific focus on studies exploring complementary versus substitution aspects of co-use. Search terms were included in Google Scholar, PsycINFO, MEDLINE, and Web of Science. Eligible studies were those that measured alcohol and cannabis co-use in human samples in laboratory, survey, or ecological momentary assessment studies, or that directly referenced substitution or complementary patterns of use. Search results returned 650 articles, with 95 meeting inclusion criteria.

DISCUSSION AND CONCLUSIONS: Results of this review reveal compelling evidence for both substitution and complementary effects, suggesting nuanced yet significant distinctions across different populations examined in these studies. Several mechanisms for the impact of cannabis use on alcohol-related outcomes are identified, including patterns and context of co-use, timing and order of use, cannabinoid formulation, pharmacokinetic interactions, and user characteristics (including diagnostic status), all of which may influence substitution versus complementary effects. This review will inform future research studies examining this topic in both clinical and community samples and aid in the development of treatment and prevention efforts targeting those populations most vulnerable to negative consequences of co-use. Finally, this review highlights the need for additional research in more diverse samples and the use of mixed-methods designs to examine both pharmacological and contextual influences on co-use.