First tobacco product tried among adolescents based on race/ethnicity and socioeconomic status

Abstract

In recent years, the tobacco use landscape among youth has changed greatly, underscoring a need to understand current tobacco product initiation patterns. This study aimed to examine if adolescents’ first tobacco product tried differs by race/ethnicity and socioeconomic status (SES). Cross-sectional survey data were collected from 6 Connecticut high schools in Spring 2019 (N = 4875). The analytic sample comprised ever users of tobacco products who reported on the first tobacco product they tried (N = 2530; 53.6% female). Of ever tobacco product users, 64.6% reported that their first tobacco product tried was an e-cigarette, 24.4% reported blunts, 6.5% reported another tobacco product, and 4.5% reported cigarettes. A multinomial logistic regression model examined race/ethnicity and SES as predictors of the first tobacco product tried with cigarettes as the reference group. Compared to Non-Hispanic (NH) White students, NH Black students (OR = 0.38, 95%CI = 0.17–0.82), Hispanic students (OR = 0.44, 95%CI = 0.27–0.71), and students of other NH races (OR = 0.30, 95%CI = 0.18–0.52) were less likely to report initiating tobacco product use with e-cigarettes. Additionally, compared to NH White students, NH Black students were more likely to report using blunts as their first tobacco product (OR = 3.01, 95%CI = 1.38–6.56). Compared to low SES, middle SES (OR = 1.62, 95%CI = 1.04–2.52) was associated with greater odds of reporting e-cigarettes as the first tobacco product. These findings suggest that youth of different backgrounds initiate tobacco use with different tobacco products and tailored prevention strategies are needed.

1. Introduction

In recent years, the landscape of tobacco product use has changed drastically among adolescents. In particular, rates of e-cigarette use have risen dramatically over the past decade. Self-reported past 30-day use of e-cigarettes among high school students has risen from 1.5% in 2011 to 27.5% in 2019 and e-cigarettes are now the most common tobacco product used among adolescents from all races/ethnicities (Wang et al., 2019). Understanding sociodemographic predictors of tobacco product initiation in the current diverse tobacco product market is essential for informing tobacco prevention strategies.

Earlier studies using data from 2014 to 2015 (Ross et al., 2018) and 2017 (Kowitt et al., 2019) found that cigarettes are the most common first tobacco product tried, followed by e-cigarettes. One of these studies examined race differences in tobacco product initiation and found that Black youth were more likely than White youth to use cigars as their first tobacco product (Kowitt et al., 2019). However, these studies did not include blunt use (i.e., a cigar product wherein some or all of the tobacco filler has been removed and replaced with marijuana) (Kowitt et al., 2019; Ross et al., 2018). Blunts are important to examine because youth may be using cigars primarily as blunts (Montgomery & Mantey, 2018). In a nationally representative sample of White, Black, and Hispanic/Latino adolescents, 12.0% of adolescents reported ever use of blunts, and Black adolescents reported using blunts at a higher rate than other races/ethnicities (Montgomery & Mantey, 2018). Blunt users are not only exposed to marijuana, but also to nicotine and tobacco through the blunt wrapper and any remaining tobacco filler (Peters, Schauer, Rosenberry, & Pickworth, 2016). This dual exposure may be particularly harmful as it has been shown that co-use of marijuana and tobacco increases symptoms of cannabis dependence, while also making it more difficult to quit tobacco use (Schauer, Rosenberry, & Peters, 2017).

Traditionally, low SES has been associated with higher rates of cigarette initiation among adolescents (Hiscock, Bauld, Amos, Fidler, & Munafò, 2012). However, with the changing tobacco product landscape, recent results have been more varied. There is some evidence that low SES is associated with higher rates of use of e-cigarettes, cigarettes, cigars, and smokeless tobacco (Simon et al., 2017). However, one recent study showed that tobacco product initiation via e-cigarettes was negatively correlated with free school lunch, a proxy for low SES (Kowitt et al., 2019). Given the rapidly changing landscape, further examination of SES and tobacco product initiation is needed.

In the current study, we examined the first tobacco product tried among adolescents and whether this differed by race/ethnicity or SES. Based on recent increases in e-cigarette use (Wang et al., 2019), we hypothesized that adolescents most commonly would report first trying e-cigarettes. Furthermore, based on patterns of ever use, we expected that Black adolescents would initiate tobacco products with blunts at a higher rate relative to other races/ethnicities (Montgomery & Mantey, 2018). We did not have a hypothesis about SES and first tobacco product tried given the varied findings in regards to SES and tobacco product initiation and the changing use patterns among adolescents.

2. Methods

2.1. Participants and procedures

Twenty-minute, anonymous, web-based, school-wide surveys assessing tobacco product use were administered during health or English classes in six Connecticut high schools (grades 9–12) during Spring 2019 (N = 4875). The schools were from different District Reference Groups, a classification system where Connecticut school districts with students of similar SES are grouped together for the purpose of comparison (CT Department of Education. District Reference Groups, 2006). Parents were notified through email and instructed to contact the school if they did not want their child to participate (declined participation: n = 3). The average response rate as a percentage of school enrollment was 77.5% (range 68.3%–82.0%). Prior to survey administration, research staff informed students that participation was voluntary and anonymous, and survey completion was considered assent/consent. Students received a stylus for participating in the survey. The Yale School of Medicine Institutional Review Board and the participating high schools approved the study.

2.2. Measures

2.2.1. Demographics

Participants reported race, ethnicity, sex at birth (male or female), and age. In order to classify race, students were asked, “How would you describe your race?” (White, Black or African American, Asian, American Indian or Alaska Native, Native Hawaiian or Pacific Islander, Middle Eastern, Other Race [please specify]), and Hispanic ethnicity was determined by a response of “yes” to “Are you of Hispanic, Latino, or of Spanish origin?”. Students who exclusively endorsed “White” were coded as Non-Hispanic (NH) White, those who exclusively endorsed “Black” were coded as NH Black, and all other answer race selections were coded as “Other Race” (i.e., NH multiracial [11.3%], NH Asian [2.1%], NH Native American [0.1%], NH Pacific Islander [0.2%], NH Middle Eastern [0.7%], and Other race [0.5%]).

2.2.2. Socioeconomic status

SES was measured using the 4-item Family Affluence Scale (FAS), a validated measure of adolescents’ family SES (Boyce, Torsheim, Currie, & Zambon, 2006). Sum scores were segmented into tertiles reflecting low (0–5), middle (6–7), and high (8–9) SES (Boyce et al., 2006; Currie et al., 2008).

2.2.3. Ever tobacco users

Adolescents were asked “Which of the following products have you ever tried”: disposable e-cigarette; e-hookah; cig-a-like; vape pen or EGO; Juul; any pod system other than a Juul, such as a Phix or Suorin; mods or advanced personal vaporizers; heat-not-burn product; cigarette; hookah; cigarillo or little cigar without marijuana; large cigar without marijuana; blunt [cigar, cigarillo, little cigar, or blunt wrap filled with marijuana]; and smokeless tobacco (responses = no/yes). Pictures of each product were provided. Adolescents who indicated use of any product were classified as ever users (N = 2542; 52.1% of the total sample).

2.2.4. First tobacco product tried

Ever users were asked “Which is the first product that you ever tried?” followed by a list of options with pictures of all products they previously indicated ever using.

2.3. Data analytic plan

The analytic sample was ever tobacco users who reported first tobacco product tried (N = 2530; 51.9% of the total sample). To assess whether first tobacco product tried differed by SES and race/ethnicity, a multinomial logistic regression was conducted using SPSS version 26 (IBM SPSS Statistics, 2019). We categorized first tobacco product tried as e-cigarettes (disposable e-cigarette; e-hookah; cig-a-like; vape pen or EGO; Juul; any pod system other than a Juul, such as Phix or Suorin; mods or advanced personal vaporizers), blunts, and other tobacco products with cigarettes as the reference group. Due to the low endorsement of initiation with other tobacco products (6.5%), we grouped these products together (i.e. hookah [4.5%], large cigar without marijuana [0.9%], cigarillo or little cigar without marijuana [0.6%], smokeless tobacco [0.4%], heat-not-burn [0.1%]).

Covariates were sex and age. No significant interactions were found between race/ethnicity and SES, so interaction terms were excluded from the final model.

3. Results

Descriptive statistics for the analytic sample are presented in Table 1. The sample was 53.6% female and on average 16.2 years old (SD = 1.3). The most frequently endorsed first tobacco product tried was e-cigarettes, followed by blunts, other tobacco products, and cigarettes.

Table 1.

Demographics of Ever Tobacco Users Reporting First Tobacco Product Tried (N = 2530).

Demographics
Race/Ethnicity: %
NHa White	45.4%
NH Black	8.8%
Hispanic	30.9%
Other Race	14.9%
SES:b %
High	25.5%
Middle	40.5%
Low	33.9%
Age (Mean, SDc):	16.2, 1.3
Female: %	53.6%
First Tobacco Product Tried: %
E-Cigarettes	64.6%
Blunts	24.4%
Other Tobacco	6.5%
Cigarettes	4.5%

Participants are adolescents in Connecticut who were surveyed in 2019.

Of our analytic sample of ever tobacco users (n = 2542), only 12 were missing cases. This was a small enough number that we dropped the missing cases from the final analysis.

^aNH = Non-Hispanic

^bSES = Socioeconomic Status

^cSD = Standard Deviation

The multinomial logistic regression analysis (Table 2) showed that NH Black, Hispanic, and NH students of Other Race (vs. NH White) were all less likely to report tobacco product initiation via e-cigarettes compared to cigarettes. Additionally, NH Black students were more likely than NH White students to report trying blunts (vs. cigarettes) as their first tobacco product. Middle SES students were more likely than low SES students to report trying e-cigarettes (vs. cigarettes) as their first tobacco product.

Table 2.

Associations for First Tobacco Product Tried by Race/Ethnicity and SES

	First Tobacco Product Tried				Adjusted ORc
	%				(95% Confidence Interval)
					E-cigarette	Blunt	Other
	E-cigarette	Blunt	Other	Cigarette	(v. Cigarette)	(v. Cigarette)	(v. Cigarette)
Race/Ethnicity
NHa White	79.2	13.9	4.0	3.0	Refd	Ref	Ref
NH Black	34.7	55.9	5.4	4.1	0.38 (0.17–0.82)	3.01 (1.38–6.56)	1.01 (0.38–2.70)
Hispanic	56.2	29.6	8.5	5.8	0.44 (0.27–0.71)	1.13 (0.68–1.89)	1.12 (0.61–2.07)
Other Race	55.1	27.1	10.9	6.9	0.30 (0.18–0.52)	0.86 (0.49–1.54)	1.18 (0.60–2.31)
SESb
Low SES	53.8	31.4	8.6	6.2	Ref	Ref	Ref
Middle SES	67.1	23.4	5.7	3.8	1.62 (1.04–2.52)	1.25 (0.79–1.98)	1.09 (0.63–1.88)
High SES	74.8	16.7	5.1	3.4	1.67 (0.98–2.89)	1.04 (0.59–1.85)	1.10 (0.56–2.15)

Note: Bold indicates p-value <0.05. Covariates were sex and age.

Participants are adolescents in Connecticut who were surveyed in 2019.

^aNH = Non-Hispanic

^bSES = Socioeconomic Status

^cOR = Odds Ratio

^dRef = Reference group

4. Discussion

We examined first tobacco product tried among adolescents by both race/ethnicity and SES. Within the total sample, we observed that the most commonly endorsed first tobacco product tried was e-cigarettes (64.6%), followed by blunts (24.4%), other tobacco products (6.5%), and cigarettes (4.5%). White adolescents initiated with e-cigarettes at a significantly higher rate than any other race and Hispanic ethnicity, and Black adolescents initiated tobacco use with blunts at a significantly higher rate than White adolescents, consistent with findings related to race/ethnicity differences in rates of current use of these products (Schauer et al., 2017; Wang et al., 2019). One potential explanation for the race differences in tobacco initiation with blunts may be the disproportionate marketing of cigar products to Black adolescents; neighborhoods with higher proportions of Black Americans have more advertisements for little cigars and cigarillos and sell them at lower prices; (Cantrell et al., 2013) thus, Black adolescents may have greater exposure to marketing for these products.

In regard to SES, we found that middle SES students were more likely to report initiating tobacco product use with e-cigarettes than low SES students. A potential reason for the observed association may be greater exposure to e-cigarette marketing among higher SES youth (Simon et al., 2018). Prior findings regarding the relationship between SES and e-cigarette use have been mixed, with some studies showing that e-cigarette use is associated with lower SES (Simon et al., 2017), some finding no association (Kinnunen, Ollila, Minkkinen, Lindfors, & Rimpelä, 2018), and others finding that e-cigarette awareness and trying e-cigarettes are associated with high SES (Adkison et al., 2013). However, given these conflicting results further research is required to clarify this relationship.

Two prior studies found that cigarettes were the most common first tobacco product tried among adolescents (Kowitt et al., 2019; Ross et al., 2018). However, in the current study only 4.5% of students initiated tobacco use with cigarettes. This difference likely is due to recent increases in e-cigarette use among youth: between 2017 and 2018 alone, current e-cigarette use among adolescents increased by 78% (Wang et al., 2019), a trend which has corresponded with the proliferation of the JUUL e-cigarette and other e-cigarette devices (Krishnan-Sarin et al., 2019).

Given the high rate at which adolescents initiated tobacco use with e-cigarettes, it is important that prevention efforts are focused on e-cigarettes. This is of particular importance considering that e-cigarette use is associated with a higher risk of cigarette initiation among adolescents (Barrington-Trimis et al., 2015; Bold, Kong, Cavallo, Camenga, & Krishnan-Sarin, 2016). Additionally, our data suggest that new prevention efforts should consider the role of blunts in tobacco product initiation, especially among Black adolescents. Blunts are made from cigars, and even when the tobacco is removed the wrapper continues to hold up to 18% of the nicotine content of the cigar (Peters et al., 2016). Blunt use in adolescents is particularly concerning because initiating tobacco products with blunts is associated with subsequent combustible tobacco product use (Mayer et al., 2019). Furthermore, blunt use is a form of co-use of tobacco and marijuana, and co-use is associated with increased risky health behaviors, increased mental health problems, additive negative respiratory effects, and higher levels of dependence on both tobacco and marijuana than with either substance used independently (Schauer et al., 2017). Our data also emphasize that including blunt use is important for understanding first tobacco product use, particularly among Black youth, so future studies also should include blunts in assessments of tobacco product use.

This study has some limitations. The sample was comprised of students in Connecticut, and may not be generalizable to the larger US population. Our sample also had a relatively small number of Black adolescents, which may limit statistical power. Given a rapidly changing landscape, future studies should continue to examine sociodemographic factors in tobacco product initiation over time.

This study also has several strengths; the current study provides an update on adolescent tobacco product initiation, which is necessary given ongoing changes in the tobacco product use patterns among youth (Wang et al., 2019). It also places an emphasis on race/ethnicity and SES differences in tobacco product initiation, which is essential in building a comprehensive picture of how youth are introduced to tobacco products. Finally, the current study includes blunts, which represents more accurate picture of the current tobacco product landscape.

5. Conclusions

The results of this study further demonstrate the importance of examining e-cigarettes and blunts and as well as taking sociodemographic factors into account to build a comprehensive picture of adolescent tobacco product initiation. Our findings suggest that prevention efforts focused on youth should consider vulnerable adolescent populations who are more prone to initiate with certain types of tobacco products. Furthermore, future studies should continue to examine how trends in first tobacco product use evolve over time and how product of initiation influences later tobacco use trajectories.

Highlights.

• Examining tobacco initiation via sociodemographic factors can inform prevention.

• 64.6% of adolescents first tried e-cigarettes, while 24.4% first tried blunts.

• White adolescents were more likely than other races to initiate with e-cigarettes.

• Black adolescents were more likely than White adolescents to initiate with blunts.

• Adolescents with middle SES were more likely to initiate with e-cigarettes.

Abbreviations:

NH

Non-Hispanic

SES

Socioeconomic Status

OR

Odds Ratio

Ref

Reference group

FAS

Family Affluence Scale