Tobacco and marijuana are two of the most frequently used addictive substances in the USA and worldwide (Center for Behavioral Health Statistics and Quality, 2014). According to data from the 2013 National Survey on Drug Use and Health (NSDUH), 71.7% of adults aged 18 years or older had ever used tobacco, and 27.3% reported past 30-day use (Center for Behavioral Health Statistics and Quality, 2014). In the same survey, 46.6% of adults reported ever use of marijuana, and 7.6% reported past 30-day marijuana use (Center for Behavioral Health Statistics and Quality, 2014).

Adults who use tobacco are more likely to have ever used marijuana and those who have ever used marijuana are more likely to use tobacco (Peters et al., 2012; Ramo et al., 2013). Use of tobacco and marijuana together may increase the risk of tobacco dependence and adverse health effects (Baggio et al., 2014). With legalization of medical and recreational marijuana use in several states, it has become increasingly important to identify how users consume marijuana, and how that use varies with tobacco use status in order to inform public health programs and policies.

Modes of ever marijuana use may differ between current, former, and never tobacco users. Adults who have ever used marijuana in a particular way may be more likely to use marijuana in that way in the future. Further, knowing how marijuana has ever been used among current, former, and never tobacco users provides insight into the relationship between marijuana and tobacco use. This can inform tailored public health programs and policies aimed at preventing and reducing marijuana and tobacco use in these specific groups. The purpose of this study is to estimate the proportion of modes of ever marijuana use among current, former, and never tobacco users aged 18 years and older, using Styles, an online consumer panel survey conducted among US adults in 2014.

More than half of current (56.6%) and former tobacco users (50.9%) had ever used marijuana, whereas only 13.0% of never tobacco users had ever used marijuana (Table 1). This pattern of greater ever marijuana use among current and former tobacco users compared to never tobacco users was observed across age, sex, racial/ethnic, education, and income categories. Joints were the most common mode of ever using marijuana regardless of tobacco use status.

Among ever marijuana users, 86.4% of current tobacco users, 92.5% of former tobacco users, and 79.8% of never tobacco users had ever smoked marijuana in a joint (Table 2). A higher proportion of current tobacco users (36.6%) had ever used a blunt (cigar with marijuana in it), compared to former (20.9%) and never (11.0%) tobacco users. More than half of current and former tobacco users had ever used marijuana in bowls or small glass pipes (54.2% and 50.1%, respectively) and in bongs, water pipes, or hookahs (54.9% and 51.8%, respectively). In contrast, only 27.0% of never tobacco users who had ever used marijuana had used it in a bowl or small glass pipe and 30.3% had ever used marijuana in a bong, water pipe, or hookah. Use of edible marijuana was also significantly higher in current and former tobacco users (32.2% and 30.3%, respectively) than in never tobacco users (20.3%). Ever use of marijuana in a vaporizer or other electronic device was similar among current (12.3%) and former (9.3%) tobacco users. One in ten (10.8%) current tobacco users and 8.0% of former tobacco users had ever used marijuana with heated hashish oil, hash oil, or honey butane oil. Ever using marijuana in some other way was reported by 5.8% of current tobacco users and 6.4% of former tobacco users.

Findings from this survey of US adults suggest that about half of current and former tobacco users had used marijuana at least once in their lifetime. In contrast, only 13% of never tobacco users had ever used marijuana, with similar use patterns across demographic subcategories. Joints were the most commonly reported mode of ever using marijuana, regardless of tobacco use status. However, prevalence of all modes of ever marijuana use was significantly higher in current and former tobacco users compared to never tobacco users. Additionally, nearly one in ten current and former tobacco users had ever used marijuana in a vaporizer or other electronic device. With the increase in use of electronic nicotine delivery systems (McMillen et al., 2014; Regan et al., 2013), use of these devices to consume marijuana might also increase (Malouff et al., 2013; Richter & Levy, 2014).

Our findings of higher proportions of ever marijuana use in current and former tobacco users compared to never tobacco users are consistent with previous research that indicate tobacco and marijuana use are closely associated behaviors (Agrawal et al., 2012; Rabin & George, 2015; Ramo et al., 2012; Richter et al., 2005). There are many possible explanations for this association. For example, the primary mode of use for both tobacco and marijuana is through combustion, and shared modes of use may promote this association (Agrawal et al., 2012; Agrawal & Lynskey, 2009; Baggio et al., 2014). Additionally, this association may also be influenced by shared genetic factors (Agrawal et al., 2012; Agrawal et al., 2010), pharmacological interactions (nicotine augments and prolongs the effects of marijuana) (Cooper & Haney, 2009; Tullis et al., 2003), and environmental factors (e.g., peer influences, substance availability, opportunity to use, and social norms) (Agrawal et al., 2012). Finally, the gateway drug theory, which proposes that use of a gateway drug, typically a legal drug-like tobacco, leads to the subsequent use of a usually illegal drug, like marijuana (Degenhardt et al., 2010). Recent evidence has documented a less common but growing reverse gateway pattern, with marijuana use preceding tobacco use (Agrawal et al., 2012; Patton et al., 2005; Peters et al., 2012).

Notably, 11% of never tobacco users reported using blunts (cigars with marijuana in them). Blunt use is not consistently considered as cigar or tobacco use (Delnevo et al., 2011). Some argue that blunt use is distinct from cigar use because all or most of the tobacco from inside the cigar is removed to create a blunt (Golub et al., 2006; Soldz et al., 2003), whereas others consider blunt use to be cigar use, because there is nicotine exposure from the cigar wrapper alone (Delnevo et al., 2011; Dunlap et al., 2005; Henningfield et al., 1999; Timberlake, 2009). Our results indicate that at least some ever blunt users did not consider blunt use to be cigar or other tobacco use. More research investigating how adults conceptualize blunt use and tobacco use in various contexts could inform surveillance to provide more accurate categorization of tobacco and marijuana use.

Findings from this study are subject to at least four limitations. First, generalizability may be limited because the sampling approaches used were not completely random. However, data were weighted to match 2014 U.S. Current Population Survey proportions on gender, age, household income, race/ethnicity, household size, education, census region, metro status, and prior Internet access. Second, findings might be subject to nonresponse bias if nonresponders were significantly different than responders in their tobacco or marijuana use. Third, survey responses were self-reported, which could lead to reporting bias; however, previous research has confirmed the validity of self-reported tobacco smoking (Caraballo et al., 2004). Finally, current marijuana use could not be examined due to small sample size. It will be important for future studies to use a more robust sample to further assess current marijuana use by tobacco use groups.

Despite these limitations, this study is the first to explore differences in modes of ever marijuana use between current, former, and never tobacco users. These findings confirm that tobacco and marijuana use are closely related behaviors and provide greater insight into the relationship between tobacco and marijuana use by detailing how tobacco users have ever used marijuana. Differences in modes of ever marijuana user by tobacco use status highlight the importance of considering the association between marijuana and tobacco use while developing interventions to prevent and reduce use of both substances. Greater understanding of how marijuana use differs between tobacco users and non-tobacco users could inform public health programs and policies aimed at controlling tobacco and marijuana use.