The present study addresses this gap by using the Tobacco Attitudes and Beliefs Survey II to investigate the relationship between marijuana use and PPR use among current cigarette smokers. This study examines 1) the likelihood of PPR use by marijuana use and 2) the frequency of marijuana use and current PPR use. Findings may help elucidate whether marijuana use is associated with PPR use, and if so, whether marijuana is used as a substitute or complement to PPR use. This is a cross sectional analysis of data from the TABS II, a web-based longitudinal survey of U.S. adult former and current cigarette smokers, aged 24 years old and older. The survey included topics such as individuals’ use of tobacco, tobacco-related products, marijuana, and other substances including PPRs. The present analysis used demographic data from Wave 1 from August 2015 . Wave 3 data were collected in August 2016 and included survey items on marijuana use and new items on PPR use . Surveys were administered by Qualtrics, which uses a combination of online panels to establish national samples from which survey participants can be randomly selected. Qualtrics invited potential participants to take the survey via an email notification and offered them a $10 incentive to complete each survey wave. For Wave 1, 2,378 individuals clicked on the survey link,vertical grow rack and 819 went on to complete the survey, yielding a completion rate of 34.4% . Current smokers were included in the current analysis.
The TABS II project was approved by the UCSF Institutional Review Board. Cigarette use.—Participants were categorized as a current cigarette smoker if they responded “yes” to the question, “Are you a current cigarette smoker?” and if they responded with any number of days greater than 0 for the question “During the last 7 days, on about how many days did you smoke cigarettes, even 1 or 2 puffs” or to the question “During the last 7 days, on about how many days did you smoke menthol cigarettes, even 1 or two puffs.” The question “On average, how many cigarettes a day do you smoke?” was used to control for cigarette consumption in the analysis of the relationship between cannabis use and PPR use. Marijuana use.—Definitions of each user type were: “never users,” never used marijuana in their lifetime; “ever” users, used marijuana at least once in their lifetime, but not in the past 30 days; and “current” users, used marijuana in the past 30 days. If participants answered the question “During the last 30 days, on about how many days did you use marijuana, even 1 or 2 puffs?” with any number above 0, they were classified as a current user. If participants responded with “I have never tried marijuana” to the question asking “Which of the following forms of marijuana have you EVER used?” then they were classified as a never user. If they responded to this question with any other option besides “Don’t know/refused” and if they were not categorized as a current user, they were classified as an ever user. For the analysis involving frequency of marijuana use as a continuous variable, responses to the question “During the last 30 days, on about how many days did you use marijuana, even 1 or 2 puffs?” were used. Medical marijuana lawstatus in state of residence.—Participants were categorized as: 1) no legal medical marijuana in state of residence, 2) legal medical marijuana for less than 10 years in state of residence, or 3) legal medical marijuana for 10 or more years in state of residence. PPR use.—PPR users were categorized as “never” users if they reported they had never used PPRs in their lifetime, as “ever” users if they had used PPRs but not in the past 30 days, and “current” users if they had used PPRs in the past 30 days.
If participants selected “Prescription pain relievers” for the following two questions, they were classified as current PPR users: 1) “Have you EVER used any of the following substances? Mark all that apply” and 2) “Have you used any of the following substances in the PAST 30 DAYS? Mark all that apply.” If participants selected “Prescription pain relievers” in the first question , but did not select them in the second question , then they were classified as ever users. If they did not select “Prescription pain relievers” in the question inquiring about ever use, they were classified as never users. Statistical Analysis—Descriptive analyses were used to test for normality. Chi-square tests were used for categorical variables , and an ANOVA was used for the continuous variables to compare sample characteristics between marijuana never users, ever users, and current users. For PPR status, a Bonferroni adjustment was made to account for multiple comparisons. Logistic regression was used to investigate the likelihood of PPR use in the past 30 days according to marijuana use . A logistic regression was used to examine whether the frequency of marijuana use influenced PPR use among current marijuana users. SAS University Edition, which contains SAS Studio 3.6 and SAS 9.4, was used for all analyses. Results suggest that adult current cigarette smokers have differential use of PPRs depending on their use of marijuana. Those who were current and ever marijuana users were over 2–3 times more likely to have used PPRs in the past 30 days, respectively, when compared to cigarette smokers who never used marijuana. Results support the findings of previous studies that addressed a possible complementary effect of marijuana use with PPR use. Novak, Peiper, and Zarkin analyzed NSDUH data in 2003 and 2013 and found that greater marijuana use was associated with more frequent PPR use. An analysis of NESARC data found higher levels of marijuana and cigarette use predicted initiation, re-initiation, and sustained opioid use ; and another study using NESARC data determined that marijuana use was associated with an elevated risk of using non-medical prescription opioids three years later . Two Swedish teams found similar results. In a re-analysis of a Swedish national household survey, non-medical PPR use was associated with both frequent cigarette smoking and marijuana use .
Studies with adolescent and young adult samples found non-medical use of PPRs is associated with marijuana use . Though longitudinal studies are needed to make definitive conclusions about the nature of the relationship between marijuana and PPR use among cigarette smokers, the interface among biological effects of PPRs, marijuana, and nicotine could influence the strength and direction of this relationship. For one, PPRs and marijuana share anti-nociceptive effects, the two substances act on some of the same brain regions, and THC partly exerts its analgesic influence by relying on opioid receptors . Nicotine additionally interacts with the opioid system, and the systems have almost identical influences in key pleasure-sensing areas of the brain . Therefore, the behavioral responses to nicotine use and withdrawal are likely affected by the opioid system . As with marijuana and opioids, nicotine has antinociceptive actions . Consequently, the interconnected neural activity and biological effects of nicotine, marijuana,cannabis vertical farming and opioids could play a role in the relationship between PPR and marijuana use among cigarette smokers. Another explanation for the higher likelihood of current PPR use among ever and current marijuana users in cigarette smokers could be that some participants had used marijuana and/or PPRs to reduce pain symptoms. Epidemiologic and prospective cohort studies point to a relationship between smoking and chronic pain, with smokers having a greater likelihood of developing chronic pain disorders than non-smokers . And the most frequently reported reason for adult misuse of PPRs in 2015 was to alleviate physical pain . Our results do not support prior findings of a negative association between marijuana and PPR use. Boehnke, Litinas, and Clauw report that among individuals with chronic pain, use of medical marijuana was negatively associated with opioid use. Further, legalization of medical marijuana has been correlated with a drop in the number of hospitalizations attributed to opioid dependence/abuse and PPR ODs, and a decline in opioid OD mortality rates . States with medical marijuana dispensaries also report fewer PPR ODs, a reduction in PPR treatment admissions, and a decline in opioid-related deaths . However, none of these studies stratified their results by cigarette smoking status. As such, it is possible that the inclusion of only current cigarette smokers in the present study could help explain the discrepancy between the present findings and other results. Of note, studies investigating the effect of marijuana use on opioid/PPR use vary in their sample composition , use of covariates , and outcome measures . This variation in study design is likely responsible for some of the discrepant findings in the extant literature. As previous studies have not stratified their analyses by cigarette smoking status, our study provides an important and unique contribution to current evidence, and this dynamic helps to explain why our findings differ from those that found a negative relationship between marijuana and PPR use. We did not find a significant association between PPR use and frequency of marijuana use among current marijuana users.
Our findings align with those of Lucas et al. , who determined that among Canadian medical marijuana users, there was no association between frequency of marijuana use and illicit drug substitution, though this finding is attenuated because their analysis was not stratified by cigarette use status. On the other hand, our findings contrast with those of Arterberry et al. , who reported that frequency of marijuana and cigarette use was predictive of opioid use among an adult sample in the NESARC. This dissimilarity may be due to differences in study design. In the work of Arterberry et al. , frequency of marijuana use was determined by asking participants how often they used marijuana in the last year, and responses were coded on a scale from 0 to 10 . In contrast, our participants reported how often they used marijuana within the past month by indicating a specific number of days from 0–30. Second, all participants in our study were current cigarette smokers, while the NESARC sample included both cigarette smokers and non-smokers. There are several study limitations. Though the TABS II used a random sample, it is not nationally representative. All data is self-reported, and there are no biomarkers for verification of tobacco, marijuana, or PPR use. The cross-sectional nature of the analysis prevents causal inference. As there were no non-smokers included in the analysis to compare with the current smokers, further analysis of the present study is warranted. We do not have data on participants’ reason for PPR use. Some participants could have medical provider issued prescriptions for pain, yet it is possible that even patients with valid prescriptions may not actually “need” a prescription pain reliever. We did not have any information on the presence, absence, or type of pain. Studies have shown that simply asking a primary care physician for a narcotic by brand name significantly increases the likelihood of being prescribed a medication and being prescribed a strong narcotic . Because pain is subjective, a definitive conclusion as to what level pain a patient is experiencing is not possible. Therefore, the only data available in a large data base is whether or not the participant had a prescription for use. We also did not measure the frequency of PPR use within the past 30 days, as we did with marijuana, nor did we ask respondents about the type of PPRs they used. In the future, questions about these details of PPR use would be beneficial to include, as they would enable a more nuanced analysis of the data. Concerning our frequency analysis, it is possible that our sample of current marijuana users was too small to capture a significant effect between frequency of marijuana use and PPR use. Finally, despite random sampling, our sample was predominantly Caucasian and older, limiting generalizability. To frame conclusions about the presence of a complementary effect between marijuana and PPR use and to identify a potential causal relationship between use of these two substances, future studies should be longitudinal, with larger and more diverse samples that include both smokers and non-smokers. The incorporation of unique aspects of MMLs into future models would be useful to more accurately determine the effects of such laws.