Trends and Factors Related to Blunt Use in Middle and High School Students, 2010–2020

Data were drawn from the 2010–2020 FYTS, a statewide cross-sectional, school-based, pencil-and-paper questionnaire administered to public middle and high school students in 67 Florida counties. Details for FYTS are available elsewhere.^39–41  The analytical sample in this study was limited to adolescents (aged 9–21 years) who answered questions about blunt use (N = 461 706 for trends analyses during 2010–2020; N = 33 038 for multivariable analyses during 2019–2020). Institutional review board approval was not required because of the deidentified nature of the publicly available data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

The definition of blunt use (as the outcome of the study) was explained to the participants as follows: “Sometimes people take tobacco out of a cigar and replace it with marijuana. This is sometimes called a ‘Blunt.’” Next, participants were asked, “Have you ever smoked part or all of a blunt?” Responses to this question were yes or no. Those who answered yes but reported not using blunts in the past 30 days were defined as past users, and those who answered yes and reported using blunts on “≥1 day during the past 30-day” preceding the survey were classified as current users. Therefore, the outcome variables for this study were defined as “past” use (“No; non-users” coded as “0” and “Yes” coded as “1”) and “current” use (“No; non-current users” coded as “0” and “Yes” coded as“1”). The FYTS included the blunt-associated questions in the school surveys starting in 2010.

Demographic characteristics include age (years), sex, race and/or ethnicity (non-Hispanic [NH] white, NH Black, Hispanic, and NH other), school level (middle or high schooler), and residential status (metro versus nonmetro or rural). County-level metropolitan status associated with the sampled schools, developed by the US Department of Agriculture Economic Research Service, was dichotomized into the metro (code 1–3) and nonmetro or rural (code 4–9, which includes completely rural counties).^41,42  We defined asthma status using the following questions. “Has a doctor or nurse ever told you that you have asthma?” Participants who responded yes to this question were considered as having “lifetime asthma.” Those who answered yes to the question “Do you still have asthma?” were considered having “current asthma.”^41,43  Those who answered “don’t know, no, or not sure” about whether they still have asthma were considered as not having asthma.

Past and current (past 30-day) use of 5 tobacco and/or nicotine products was assessed. These products were cigarettes, e-cigarettes, hookah (ie, waterpipe), cigars (cigars, little cigars, and cigarillos, ie, nonblunt cigars), and smokeless tobacco (SLT) (chewing tobacco, snuff, dip, snus, and dissolvable tobacco). For each of these 5 products, participants were asked, “Have you ever smoked/used part or all of a (name of product)?” Responses to this question were yes or no. Those who answered yes but indicated not using the product in the past 30 days were defined as past users, whereas those who answered yes and reported use of the product on “≥1 day during the past 30-day” were classified as current users. Ever use of e-cigarettes to vape marijuana oil was determined by asking “Have you ever used an electronic vapor product with marijuana oil or hash oil?” Responses to this question were yes or no, with a yes response indicating ever use.

All analyses were weighted by using SAS/STAT version 14.2 (SAS Institute, Inc, Cary, NC) survey procedures (PROC SURVEYFREQ and PROC SURVEYLOGISTIC) to account for survey design features, to offset nonresponse bias, and to be representative of middle and high schoolers in Florida. To examine the factors associated with past and current blunt use, we pooled data from the 2019 and 2020 FYTSs. Univariate differences across levels of each independent variable (demographic characteristics and tobacco use) were examined by Rao-Scott χ² tests for middle and high schoolers. We applied weighted multivariable regression models adjusted for demographic and tobacco use variables for middle and high schoolers to estimate adjusted odds ratios (aORs) and 95% confidence intervals (CIs) of past and current blunt use. All associations were considered significant at the α level of .05. We grouped our findings by school level (middle versus high schoolers) to provide a clear picture of risk differences by age group and to recommend school-based programs to prevent blunts and tobacco and/or nicotine product use.

For the time trends analysis of the annual estimated prevalence of blunt use from 2010 to 2020, we used the Joinpoint Regression software version 4.8.0.1 released by the US National Cancer Institute.⁴⁴  Joinpoint regression enables the user to test whether an apparent positive or negative change in trend is statistically significant by breaking the data into time segments. For each time segment, we calculated the annual percentage change (APC) in past and current blunt use during that period and determined if the APC was statistically different from 0 (no trend).⁴⁴  Joinpoint regression analyses were stratified by sex, race, and school level, providing the SE for each annual prevalence estimate by using PROC SURVEYFREQ software.

We examined potential modification (interaction) effects of sex and race and/or ethnicity on outcomes of interest (current and past blunt use) in multivariable regression models (controlling for age and residence status) and stratified trends in blunt use by sex and race (see Supplemental Tables 6 and 7). We also compared statewide blunt use with nationwide marijuana use using published data from the US YRBS (2011–2019) among high school students.⁴ 

Of the 461 706 respondents during 2010–2020, 50.7% were male, 56.9% high schoolers, 41.2% NH white, 22.1% NH Black, 30.6% Hispanic, and 6.2% NH other. As displayed in Supplemental Table 3, past blunt use for 11 waves (2010–2020) ranged from 19.6% to 20.0%, and current blunt use for the same period ranged from 10.5% to 12.7% (Supplemental Table 4). As shown in Supplemental Fig 2, the past and current use of blunt in Florida high schoolers had a similar pattern as marijuana use in US high schoolers from 2011 to 2019.

As displayed in Fig 1 and Supplemental Table 5, whereas the prevalence of past (APC = −5.32%; 95% CI, −7.9% to −2.7%) and current (APC = −5.28%; 95% CI, −7.1% to −3.4%) blunt use significantly decreased from 2010 to 2015, increasing trends in current use prevalence (APC = 14.91%; 95% CI, 0.5% to 31.4%) were observed from 2015 to 2018 and has been approximately constant ever since. Similar increasing trends were observed in current blunt use among female students (APC = 14.92%; 95% CI, 1.1% to 30.7%), middle schoolers (19.57%; 95% CI, 2.3% to 39.8%), and NH white students (APC = 11.12%; 95% CI, 2.0% to 21.1%) from 2016 to 2020. Joinpoint models revealed no significant change for average APC for past and current blunt use overall and by sex, race and/or ethnicity, and school level (all P values >.05; data not shown).

In the adjusted multivariable regression model as shown in Table 1, those who were past blunt users (versus nonusers) were more likely to be older (aOR = 1.32; 95% CI, 1.17 to 1.49), NH Black (versus NH white; aOR = 2.52; 95% CI, 1.76 to 3.62), past users of cigars (aOR = 5.23; 95% CI, 2.93 to 9.34), cigarettes (aOR = 2.14; 95% CI, 1.45 to 3.14), e-cigarettes (aOR = 8.01; 95% CI, 5.73 to 11.19), and hookah (aOR = 2.58; 95% CI, 1.40 to 4.73), current users of e-cigarettes (aOR = 1.50; 95% CI, 1.07 to 2.12), and ever users of e-cigarettes to vape marijuana oil (aOR = 8.56; 95% CI, 6.17 to 11.86). Those who were current blunt users (versus noncurrent users) were more likely to be older (aOR = 1.22; 95% CI, 1.07 to 1.39), NH Black (versus NH white; aOR = 2.36; 95% CI, 1.41 to 3.94), Hispanic (versus NH white; aOR = 1.79; 95% CI, 1.23 to 2.59), past users of e-cigarettes (aOR = 3.63; 95% CI, 2.22 to 5.91) and hookah (aOR = 2.32; 95% CI, 1.33 to 4.07), current users of cigars (aOR = 4.93; 95% CI, 1.98 to 12.25) and e-cigarettes (aOR = 3.75; 95% CI, 2.58 to 5.45), and ever users of e-cigarettes to vape marijuana oil (aOR = 8.54; 95% CI, 5.90 to 12.35).

In the adjusted multivariable regression model as shown in Table 2, those who were past blunt users (versus nonusers) were more likely to be older (aOR = 1.15; 95% CI, 1.08 to 1.23), NH Black (versus NH white; aOR = 2.20; 95% CI, 1.76 to 2.75), past users of cigars (aOR = 3.10; 95% CI, 2.25 to 4.26), cigarettes (aOR = 1.93; 95% CI, 1.50 to 2.50), e-cigarettes (aOR = 4.52; 95%C, 3.71 to 5.50), and hookah (aOR = 3.00; 95% CI, 2.20 to 4.73), current users of e-cigarettes (aOR = 1.57; 95% CI, 1.27 to 1.93), and ever users of e-cigarettes to vape marijuana oil (aOR = 8.32; 95% CI, 6.88 to 10.06). Those who were current blunt users (versus noncurrent users) were more likely to be NH Black (versus NH white; aOR = 1.92; 95% CI, 1.49 to 2.48), past users of e-cigarettes (aOR = 2.19; 95% CI, 1.66 to 2.90) and hookah (aOR = 2.06; 95% CI, 1.59 to 2.66), current users of cigars (aOR = 5.98; 95% CI, 3.92 to 9.13), cigarettes (aOR = 1.80; 95% CI, 1.11 to 2.91), e-cigarettes (aOR = 3.04; 95% CI, 2.49 to 3.71), and hookah (aOR = 3.62; 95% CI, 1.02 to 2.68), and ever users of e-cigarettes to vape marijuana oil (aOR = 5.63; 95% CI, 4.56 to 6.96). Past users of blunt were more likely to report having lifetime asthma (aOR = 1.27; 95% CI, 1.01 to 1.62) compared with nonusers.

As shown in Supplemental Tables 6 through 7, our analyses revealed no effect modifications between sex and race and/or ethnicity on past or current blunt use both in trends and multivariable models (all P > .05).

This study revealed that during 11 years (2010–2020), the proportion of self-reported past and current blunt use by Florida adolescents decreased 5 APC points per year in 2010–2015 (from 19% to 16% for past users and from 10.5% to 8.5% for current users) and then increased 15 APC points per year in 2015–2017 (from 8.5% to 13.6% only for current blunt use), which was more pronounced in middle schoolers than in high schoolers and remained almost constant ever since (during 2018–2020). These trends of blunt use follow the same pattern observed in trends of marijuana use among a national sample of US high schoolers from 2011 to 2019.⁴  This reversal trend in blunt use coincides with legalizing medical marijuana consumption in Florida in 2017,²⁸  which may have further facilitated accessibility for adolescents.^21,45  Another possible reason is the low perceived risk of marijuana use among adolescents in recent years, which might have increased the appeal and experimentation among adolescents.^29,46  As of March 7, 2021, 37 states and the District of Columbia have legalized the use of marijuana for medical purposes, including 12 states with retail marijuana laws (although marijuana remains illegal under the federal law).²⁸  Substantial changes in marijuana policy that include provisions for medical and/or retail use make it pivotal to monitor its use trends in any form (eg, blunt) among adolescents to inform policy makers and clinicians (particularly pediatricians) to make effective tailored interventions to curb marijuana use.

During 2019–2020, the use of blunts to consume marijuana was reported by 11.8% (estimated 344 000) of middle and high schoolers in Florida, with higher rates among NH Black students and co-users of tobacco and nicotine products. The trends of past and current blunt use were roughly parallel for NH Black and NH white adolescents, with higher prevalence of use among NH Black adolescents in recent years. Recent studies reported that being Black (compared with other races) was a strong predictor of co-use of marijuana and tobacco among young adults, which is likely because of the high prevalence of blunt use among this minority population.^21,47  In our study, even with a considerable increase in current blunt use among NH white individuals in 2016–2020, Black individuals had the highest prevalence of blunt use over 11 years (2010–2020). Several factors may drive race disparities in blunt use. African Americans may prefer blunts to joints (marijuana wrapped in a cigarette paper that does not contain nicotine) because a blunt can hold more marijuana inside it and is inexpensive, burns more slowly, and is easier to transport and conceal.^33,34,48  Moreover, cigars have been historically and heavily marketed to the Black community by the tobacco industry through targeted advertising^49,50  and sponsored music events.⁵¹  More importantly, Black individuals are more likely than other racial groups to have friends who smoke blunts.^21,33,34,48  Prevention programs to reduce blunt use among adolescents and prevent subsequent co-use of tobacco and marijuana must address the racial disparities in blunt use, especially when Black adolescents have decreased access to substance use treatment compared with white adolescents.⁵² 

Our analysis revealed that despite similar downward trends in past and current blunt use among males and females in 2010–2016, a substantial increase in current blunt use was observed among female adolescents in 2016–2020 (15 percentage points per year; from 7.3% to ∼13.0%). This trend merits serious attention, especially because female marijuana (including blunt) users are more vulnerable to mental health disorders such as depression and have a greater likelihood to develop marijuana addiction than male users.^53,54  This is concerning because results from the 2019 YRBS revealed no significant sex difference in the current use of marijuana among high schoolers (male = 22.5% versus female = 20.8%).⁴  Taken together, these findings highlight the need for a better understanding of factors associated with marijuana use among female adolescents. Such understanding will help in developing sex-specific prevention campaigns and interventions aimed at curbing blunt use among adolescents, especially in light of shifting policies on marijuana (eg, legalized recreational use).^28,46 

In addition to smoking blunts, e-cigarette vaping has become a popular way to consume tetrahydrocannabinol oil and other substances.^30,55  This notion was in agreement with our findings, which revealed that both middle and high schoolers who vaped marijuana using an e-cigarette device had >15-fold odds of being a current blunt user. This is not surprising because marijuana is the most commonly used illicit substance among adolescents in the United States,²⁷  which also coincided with a high prevalence of current e-cigarette use among middle schoolers (4.7%) and high schoolers (19.6%) in 2020.⁵⁶  In addition, we found that, among high schoolers, past blunt users were more likely to report having lifetime asthma. Although the scope of our study cannot rule out any potential likelihood of reverse causality (ie, individuals with asthma use marijuana to relieve symptoms), according to the American Thoracic Society, smoking marijuana (and blunt) can cause an asthma attack with dire consequences (eg, hospitalization and even death).²⁶  Therefore, the co-use of blunts and vaping marijuana among adolescents, as well as having lifetime asthma among past blunt users, raises a serious concern about detrimental health effects and calls for rapid regulatory actions to curb this potential epidemic among adolescents, especially in light of the recent tragic e-cigarette- or vaping-associated lung injury outbreak.³¹  Most importantly, there is an urgent need to incorporate educational elements regarding the harms of marijuana consumption among adolescents into prevention campaigns that are aimed at curbing tobacco products, especially e-cigarettes, among this vulnerable age group.

This study has some limitations. The cross-sectional nature of this study does not allow us to deduce causality or determine the temporality of the observed associations. Further research using longitudinal studies is warranted to investigate the temporality relation between blunt use and other tobacco products and whether it leads to some health issues such as asthma. Moreover, there was no information regarding the type and brand of blunt wraps used to smoke marijuana. Although traditionally, blunts are made of tobacco leaves and contain nicotine,⁵⁷  some marijuana consumers, especially those who are not tobacco and/or nicotine users, buy prerolled nicotine-free blunt wraps (eg, natural palm leaf style wraps) with a variety of available flavors.⁶  Therefore, surveys like the FYTS, YRBS, and National Youth Tobacco Survey should include a question or questions distinguishing the type of blunt wrap used and whether it contains nicotine. Finally, the findings of our study may not generalize to states beyond Florida and to adolescents who are not enrolled in school. Nonetheless, to the best of our knowledge, this is the first study used to examine the co-use of blunts and various tobacco products among a large sample of adolescents in the United States using time trend analyses.

Current use of blunts was reported by a large number of Florida students, has increased over time (especially among female students), and was particularly prevalent among NH Black individuals, and co-users of tobacco products such as cigarettes, e-cigarettes, hookah, cigars, and SLT. Our findings can inform future regulatory actions (eg, incorporating marijuana [and blunt] modules into existing tobacco and/or nicotine prevention programs) and help public health practitioners in designing school-based prevention programs (eg, educating adolescents about the harms of tetrahydrocannabinol and nicotine on the developing brain and respiratory health).^58–60  Such regulations and programs are urgently needed and can provide a substantial contribution to the health of the next generations.

We thank the Florida Department of Health for providing 2019 FYTS data. We also acknowledge all middle and high school students who participated in the FYTS.

aOR

adjusted odds ratio

APC

annual percentage change

CI

confidence intervale-cigarette, electronic cigarette

FYTS

Florida Youth Tobacco Survey

NH

non-Hispanic

SLT

smokeless tobacco

YRBS

Youth Risk Behavior Survey