There is a national trend toward statewide legalization of medical marijuana despite federal classification of marijuana as a Schedule I illicit drug. There are compelling arguments for and against medical marijuana legalization and its potential impact on an array of complex social issues . Residents in states where medical marijuana is legal are more likely to have tried marijuana, report current marijuana use, and be diagnosed with marijuana abuse or dependence . Additionally, there is preliminary evidence to suggest that there is likely a dose-response relationship between the number of years since legalization and marijuana prevalence rates . A key question regarding more liberal marijuana policies is whether and how they affect use of other drugs including addictive and harmful substances like tobacco. Previous studies have found a strong positive association between cigarette and marijuana use . Epidemiologic data indicate that the prevalence of tobacco and marijuana co-use has increased from 2003 to 2012 . Moreover, the increase in co-use occurred specifically among those ages 26–34 years, and the greatest percent increase, in those ages 50 years and older . It is unknown, however, if this national increase in co-use is directly associated with statewide legalization of medical marijuana. If marijuana policies are indeed associated with co-use, the current trend toward legalization of medical and/or recreational marijuana, without any regulatory action, has the potential to influence patterns of cigarette and marijuana use/co-use over time. An increase in cigarette and marijuana co-use has the potential to create challenges for cigarette smokers who want to quit. There is evidence to suggest that cigarette and marijuana co-use is associated with greater nicotine dependence . Possible explanations for this link include the role of the endocannoboid system in nicotine metabolism , genetic predisposition for co-use grow lights for cannabis, and various environmental and cultural influences . The relationship between co-use and nicotine dependence, however, is understudied in adults, particularly among those ages 50 years and older.

Since nicotine dependence is influenced by both nicotinic receptors and nicotine associated metablism that change with age , we can expect nicotine dependence among cigarette and marijuana co-users will also vary over the lifespan. Few studies have examined cigarette and marijuana co-use and nicotine dependence from adolescence through adulthood. As the nation is well-past the tipping point on medical marijuana legalization, studies are needed to take a closer look into whether marijuana policies have the potential to influence tobacco control efforts at the population level. For example, over time, it is likely that greater access to legal marijuana will increase the absolute number of co-users who have greater nicotine dependence and difficulty quitting cigarettes. Such data can help to identify subset populations at higher risk of nicotine dependence and could have both policy and treatment implications in tobacco control. In this study, we sought to examine relationships between medical marijuana laws and cigarette and marijuana co-use. Additionally, we examined the likelihood of nicotine dependence in co-users. We analyzed data from the 2013 National Survey on Drug Use and Health and stratified the analysis by age categories. Results from this study can inform the direction of future medical marijuana policies that may inadvertently affect tobacco control efforts.We analyzed cross-sectional data from the 2013 NSDUH conducted by the Substance Abuse Mental Health Services Administration . The primary purpose of NSDUH is to measure prevalence and correlates of drug use in the civilian, non-institutionalized U.S. population aged 12 years and older. Since 1991, NSDUH has consisted of an independent multistage area probability sampling design for each state and the District of Columbia and uses a combination of the Computer-Assisted Interviewing and Automated Computer Assisted Interviewing instruments in selected individuals and households . The survey offered $30 in cash to participants and was conducted in 2013 by Research Triangle Institute . The final survey consisted of 67,838 CAI interviews with a weighted screening response rate of 84% and an interview response rate of 72%. The public use file consisted of 55,160 records due to a sub-sampling step which included a minimum item response requirement for weighting and further analysis.

A detailed description of the questionnaire items, sampling methodology, data collection/ response rates, and sample weights is published elsewhere . The present study was exempt from the University of California San Francisco’s Human Research Protections Program approval since data were publically available and subjects cannot be identified. In this analysis, only those with complete responses for all measures were included. Additionally, while the analysis included participants aged 50–64 years, those 65 years of age and over were excluded due to a small sample size . The final sample included 51,993 participants.Descriptive statistics are reported for demographics, cigarette and marijuana use, and lifetime depression as well as chi-square tests of differences by statewide medical marijuana legalization status . One-way ANCOVA models tested for differences in marijuana use and cigarette and marijuana co-use in the overall sample, and separately for each age category, between states where medical marijuana was legal vs. illegal, adjusting for age , gender, race/ethnicity, education, age at first cigarette initiation, age at first marijuana initiation, and lifetime depression. Additionally, we calculated mean NDSS and frequency of TTFC scores by statewide legalization categories across age groups. In the overall sample and within each age category, two logistic regression models examined nicotine dependence, as measured by NDSS and TTFC scores, in cigarette and marijuana co-users . Models were adjusted for age , gender, race/ethnicity, education, lifetime depression, and statewide medical marijuana legalization status. Bonferroni adjustments were applied to all models with over five independent variables . In this analysis, we used the Taylor series method for replication methods to estimate sampling errors of estimators based on complex sample designs. The regression coefficient estimators were computed by generalized least squares estimation using element-wise regression. The procedure assumes that the regression coefficients are the same across strata and primarily sampling units . All models were run in SAS 9.4 using the SURVEY procedures to obtain weighted estimates to increase the generalizability of the findings . Findings indicate an association between statewide legalization of medical marijuana and cigarette and marijuana co-use despite lower cigarette prevalence in states where medical marijuana was legal. Co-use was particularly robust among 18–34 year olds. Overall, cousers were more likely to be nicotine dependent compared to those who did not use marijuana, and 12–17 year old adolescent and 50–64 year old adult co-users were 3-times more likely to have nicotine dependence . These data suggest that medical marijuana legalization could inadvertently affect prevalence of co-use, which is linked to greater nicotine dependence, and the potential to create more barriers to smoking cessation . As more states pass marijuana laws, and the legal marijuana industry is poised to cultivate a landscape of greater access and exposure to marijuana , it is recommended that stakeholders in tobacco control prepare for any unintended effects on tobacco use including the possibility of tobacco initiation/ reinitiation among former smokers and greater nicotine dependence in current smokers . Longitudinal research is needed to evaluate the effect of state marijuana policy on tobacco use and marijuana and tobacco co-use.

Given the nationwide increase in co-use , there may be uptake of marijuana use among cigarette users as states, change their marijuana policies and cigarettes smokers gain greater exposure and access to legal marijuana. It is possible that medical marijuana may be providing cigarette smokers with an alternative to tobacco especially as the stigma associated with tobacco continues to rise and the perceived harmfulness of marijuana decreases with legalization . Further, it might be perceived that the effects of marijuana can curb nicotine cravings and withdrawal symptoms to aid in smoking cessation . Finally, alternative tobacco products such as electronic nicotine delivery systems, which are commonly promoted as cessation aids and “safe” alternatives to smoking cigarettes , might also promote use of marijuana and THC oil with vaporizers . Co-use should therefore be monitored over time and examined in response to changes in marijuana policies that will further propel industry promotion of co-use and vaping. As expected, the prevalence of cigarette and marijuana co-use differed according to age. The positive association between medical marijuana legalization and co-use was greatest among 18–34 year olds. Previous studies with adolescents have reported greater prevalence but no increase in marijuana use or changes in permissive attitudes in states where medical marijuana was legal ,grow cannabis suggesting that greater marijuana use, and therefore greater co-use, preceded medical marijuana legalization. However, most published studies have focused only on adolescents under the age of 18 years and do not reflect the adult population to which medical marijuana policies apply . Therefore, long-term longitudinal studies are needed to monitor the effects of marijuana legalization, marijuana initiation/ re-initiation, cigarette initiation/ reinitiation, and patterns of co-use across all age categories. Additionally, it is recommended that such studies take into account statewide variables including number of years since the policy went into effect to adequately capture any measurable changes. These data are needed to explore the growing evidence and public health concerns about the potential “gateway” effect of marijuana on cigarette initiation and nicotine dependence in adolescents and young adults in addition to the potential for re-initiation of cigarettes among former tobacco users. As more states pass marijuana policies, potential increases in co-use could have important treatment implications. Cigarette smokers who also reported current marijuana use were more likely to have nicotine dependence, which is a known predictor of smoking and quitting behavior . The positive link between co-use and nicotine dependence was observed across age categories but these associations differed across measures of dependence . We analyzed both NDSS and TTFC. NDSS scores might have been a better measure of nicotine dependence in our comparison across age groups since the scale addresses five aspects of dependence . In comparison, the TTFC single-item scores might not have captured dependency, particularly in adolescent and young adult populations, who have yet to become regular and established smokers. Other studies have shown problems in using TTFC as a measure of dependence in young adults . Since our analysis included both adolescents and adults, we report both NDSS and TTFC measures of nicotine dependence. In addition, in the present study, cigarette smokers who reported ever but not current marijuana use were at greater risk of having nicotine dependence compared to never marijuana users. This finding supports that the effect of THC exposure on nicotine receptors may be irreversible . Studies are needed to further examine both short term and possibly even the long-term effects of THC and nicotine exposure on nicotine dependence and tobacco cessation. In this analysis, 12–17 year old adolescent and 50–64 year old cigarette and marijuana cousers had the highest odds of having nicotine dependence.

These findings support previous studies linking co-use and nicotine dependence in adolescents and young adults and add to preliminary data that this association was also stable in adults and, surprisingly, particularly robust in 50–64 year old adults. These findings reflect evidence of a U-shaped effect between age and nicotine dependence which peaks at age 50 years due to changes in nicotinic receptors and nicotine-associated metabolism with age , and suggest that this relationship was stable among co-users. Studies are needed to determine the extent to which THC exposure and/or current marijuana use add to this effect . Additionally, 50–64 year olds may represent a unique birth cohort who spent their formative years during the 1960’s and 1970’s with minimal tobacco regulations coupled with a counterculture that promoted marijuana use among a large population . More studies on the Baby Boomer generation, specifically, their perceptions about marijuana, current marijuana use including purpose of use , modality, cigarette co-use, and health outcomes could provide a glimpse into the future as continued legalization will likely influence social norms across the general population . As more states adopt liberal marijuana policies, more studies are needed to understand co-use including the relationship between THC and nicotine in addition to other individual-level factors such as genetics and personality traits that might influence dependence and cessation . We found higher percentages of non-Hispanic Whites and Blacks/ African-Americans in states where medical marijuana was illegal. In this study, these results may be attenuated since our analysis comparing nicotine dependence depended on exclusion of blunt use.