Symptoms of tobacco dependence among middle and high school tobacco users – Data from the 2019–2020 National Youth Tobacco Survey

Abstract

Introduction:

Previous studies have indicated that youth who use tobacco products, including cigarettes, cigars, and smokeless tobacco, demonstrate dependence symptoms. However, the tobacco marketplace has expanded dramatically in recent years, and few studies have examined dependence symptoms among youth who use novel products. This study combined 2019–2020 National Youth Tobacco Survey data to report the prevalence and determinants of tobacco dependence symptoms among U.S. middle and high school current (past 30-day) tobacco users.

Methods:

Prevalence estimates were calculated to examine dependence outcomes and other covariates by user groups (single product users and multiple product users). Multivariable logistic regression analyses were used to identify independent predictors of tobacco dependence among current users of cigarettes, cigars (regular cigars, cigarillos, and little cigars), e-cigarettes, heated tobacco products, hookah, pipe tobacco, bidis, and smokeless tobacco products (chew, snuff, dip, snus, and dissolvables).

Results:

Among current tobacco users, 15.7 % (95 % CI: 14.2–17.3) reported wanting to use tobacco within 30 min of waking and 28.3 % (95 % CI: 26.3–30.5) reported strong cravings for tobacco in the past 30 days. Nearly-two-thirds of current users were single product users, of which 80.5 % reported using e-cigarettes. Reporting of dependence symptoms was generally associated with multiple product use, higher frequency of use, earlier initiation age, and use of flavored products.

Conclusions:

Among U.S. adolescents, a considerable amount of current tobacco product users, even infrequent users, reported symptoms of dependence. These findings highlight the continued importance of prevention strategies for youth tobacco experimentation and progression to regular use.

1. Introduction

Tobacco product use is the leading cause of preventable disease and premature death in the United States; nearly all tobacco product use begins during youth and young adulthood (National Center for Chronic Disease Prevention, 2012). In 2020, 16.2 % (an estimated 4.47 million) of middle and high school students reported current tobacco use (Gentzke et al., 2020).

The U.S. tobacco marketplace contains a wide range of product types (e.g., combusted, smokeless, and electronic products) that offer varied levels of innovation, including product characteristics such as flavors and nicotine content, all of which complicate the understanding of tobacco use behaviors among youth (Mermelstein, 2014). Flavors, in particular, may mask aversive aspects of tobacco use such as the bitterness of tobacco, and harshness/irritation, thereby facilitating inhalation, contributing to appeal among current users, and increasing palatability for new users (Baker et al., 2021; Kostygina et al., 2016; Peterson et al., 2021). Evidence from national surveys of youth, young adults, and adults indicate that flavors are associated with tobacco use initiation (Ambrose et al., 2015; Nyman et al., 2016), more frequent use, and increased risk of future cigarette initiation (Barrington-Trimis et al., 2018; Chen et al., 2017; Dai & Hao, 2016; Leventhal et al., 2019). In 2019, 69.6 % (an estimated 4.31 million) of adolescent tobacco product users reported using ≥ 1 flavored tobacco product (Wang et al., 2019). However, while innovation and flavors may appeal to youth, it is the nicotine in tobacco products that causes dependence and the progression from initiation to experimentation to regular use (National Center for Chronic Disease Prevention, 2012). Further, flavors in tobacco products play a role in tobacco dependence by attracting youth to products that contain nicotine, can mask the hardness of inhalation, are listed as a reason for why youth use tobacco products, and in some cases, like for menthol in cigarettes, also interact with nicotine (King, 2022; Kostygina et al., 2016; Wickham, 2020). Given that tobacco use largely begins in adolescence and considering the availability of novel tobacco products, understanding the presence of tobacco dependence symptoms among adolescents, and examining dependence across tobacco product use patterns is critical.

In 2014, Apelberg and colleagues analyzed tobacco dependence using the 2012 National Youth Tobacco Survey (NYTS) and found increased reporting of dependence symptoms among adolescents to be independently associated with multiple product use, increased frequency of use, earlier age at first use, and being female (Apelberg et al., 2014). However, the study also noted that symptoms of tobacco dependence were reported even among adolescent tobacco users who used tobacco products infrequently (Apelberg et al., 2014). Another study found that tobacco dependence (including symptoms of cravings and withdrawal) was present among adolescents who reported infrequent smoking (O’Loughlin et al., 2003). More recently, a study using 2017–2018 NYTS data found increased reporting of tobacco dependence among frequent users (≥20 days) and multiple tobacco product users; (Gomez et al., 2020) however, this study did not assess dependence measures separately or examine flavored tobacco use.

The current study uses 2019–2020 NYTS data to update estimates of prevalence and examine characteristics of self-reported tobacco dependence reported by Apelberg et al. (2014) among all current tobacco users, single product users, and multiple product users. In addition, this paper examines whether flavored tobacco product use is associated with symptoms of tobacco dependence.

2. Materials and methods

2.1. Data source

NYTS is a cross-sectional, voluntary, school-based survey administered to U.S. middle (grades 6–8) and high school (grades 9–12) students enrolled in private or public schools. NYTS produces nationally representative estimates of adolescents’ tobacco use and associated factors. In the current study, combined data from the 2019–2020 NYTS were analyzed. The sample sizes and response rates were n = 19,018 (66.3 %) in 2019 and n = 14,531 (43.6 %) in 2020; the total analytic sample was n = 33,549. The 2020 data collection period was truncated due to the COVID-19 pandemic; however, the truncated timeline did not affect the original survey design or the national representativeness of the weighted estimates. The NYTS methodology has been described in further detail elsewhere (Centers for Disease Control and Prevention, 2020). The NYTS study protocol was approved by the institutional review board of the U.S. Centers for Disease Control and Prevention.

2.2. Measures

Demographic characteristics: In this study, school type (middle or high school), sex, and race/ethnicity were included. Race/ethnicity was recoded into four categories: non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic Other. Hispanic respondents could be of any race (White, Black or African American, or other race [American Indian or Alaska Native, Asian, Native Hawaiian or other Pacific Islander]).

Tobacco Use: Current tobacco users were characterized into single or multiple (≥2) product users. Current use was defined as use on ≥ 1 day of the past 30 days of the following products: e-cigarettes, cigarettes, cigars (regular cigars, little cigars, cigarillos), smokeless tobacco (chewing tobacco, dip, snuff, snus, dissolvables), hookah, pipe tobacco, bidis, or heated tobacco products. If a respondent reported use of ≥ 2 products in the past 30 days, frequency of use was based on the product used most frequently (1–5 days, 6–19 days, ≥20 days); age at first tobacco use was based on the earliest product used (<11 years, ≥11 years).

Flavored product use was defined as use of ≥ 1 flavored tobacco product, including menthol cigarettes, among current tobacco users. For each non-cigarette tobacco product, current users were asked, “Were any of the [tobacco product] that you used in the past 30 days flavored to taste like menthol, mint, clove or spice, alcoholic drinks, candy, fruit, chocolate, or any other flavor?” Current cigarette smokers were categorized as flavored (menthol) cigarette smokers if they reported usually smoking menthol cigarettes or indicated “Kool” or “Newport” as the usual brand smoked in the past 30 days. Current tobacco users who provided no valid responses to any flavor questions were excluded from the analyses (n = 2). Questions on flavored product use differed slightly in 2019 and 2020; the specific question wording is in the table footnotes.

Tobacco Dependence: Current tobacco users were categorized as having any symptoms of tobacco dependence if they answered “Yes” to the question, “During the past 30 days, have you had a strong craving or felt like you really needed to use a tobacco product of any kind?,” or selected “Within 5 min,” or “From 6 to 30 min” for the question, “How soon after you wake up do you want to use a tobacco product?” These dependence measurement items have been shown to be reliable and valid (Gomez et al., 2020; Nonnemaker et al., 2004; O’Loughlin et al., 2002).

2.3. Statistical analysis

Weighted prevalence estimates and 95 % CIs were calculated for demographic and product use characteristics among current tobacco users overall, single product users, and multiple product users. Multivariable logistic regression modeling was conducted among each user group and for each tobacco dependence measure (i.e., craving and time to first tobacco product), adjusting for the following covariates: sex, school type, race/ethnicity, age at first use, frequency of use, and flavored product use. Among current tobacco users, multiple product use was included as a covariate. Among single product users, the type of tobacco product used was included as a covariate. Analyses were conducted using SAS-callable SUDAAN (version 11.0.3, Research Triangle Institute) to account for the complex survey design. Estimates with a relative standard error > 30 % or an unweighted denominator < 50 were suppressed.

3. Results

In 2019–2020, the overall prevalence of current tobacco use among U.S. middle and high school students was 19.7 % (95 % CI: 18.4–21.0). The prevalence of adolescents who reported single product use was 12.6 % (95 % CI: 11.8–13.5) and multiple product use was 7.1 % (95 % CI: 6.4–7.8) (data not shown).

The majority of current tobacco users were in high school (78.3 %, 95 % CI: 74.8–81.4), non-Hispanic White (59.6 %, 95 % CI: 55.8–63.4), and single product users (64.1 %, 95 % CI: 61.8–66.4) (Table 1). Among single product users, e-cigarettes were the most used product (80.5 %, 95 % CI: 77.9–82.9), followed by cigars (6.4 %; 95 % CI: 5.1–8.1), and smokeless tobacco (4.1 %, 95 %: 3.1–5.3). The majority of single product users reported infrequent use (58.2 %, 95 % CI: 55.8–60.6). While only 9.1 % of single product users reported initiating tobacco use < 11 years, 27.3 % of multiple product users reported initiating tobacco use < 11 years. Among multiple product users, the majority were male (58.8 %, 95 % CI: 55.9–61.8), used tobacco frequently (53.3 %, 95 % CI: 50.3–56.4), and used flavored products (82.4 %, 95 % CI: 80.0–84.5). Further, 89.1 % of multiple product users reported using e-cigarettes, 50.9 % cigars, 48.4 % cigarettes, 33.8 % smokeless, and 27.5 % hookah (data not shown). Only 22.5 % of single product users reported ≥ 1 dependence symptom versus 49.0 % of multiple product users. Overall, 28.3 % of current tobacco users reported having a strong craving for a tobacco product, while 15.7 % reported desire to use tobacco within 30 min of waking.

Table 1.

Characteristics of current tobacco product users, by tobacco use status, 2019–2020 National Youth Tobacco Survey.

	Current tobacco usersa Unweighted n = 6,471 Weighted N = 5,360,000		Current single product usersb Unweighted n = 4,185 Weighted N = 3,430,000		Current multiple product usersc Unweighted n = 2,286 Weighted N = 1,920,000
	n	% (95 % CI)	n	% (95 % CI)	n	% (95 % CI)
School type
Middle school	1,598	21.7 (18.6–25.2)	1,040	21.4 (18.2–24.9)	558	22.3 (18.6–26.3)
High school	4,843	78.3 (74.8–81.4)	3,130	78.6 (75.1–81.8)	1,713	77.7 (73.7–81.4)
Sex
Male	3,434	52.6 (50.8–54.5)	2,077	49.1 (47.1–51.2)	1,357	58.8 (55.9–61.8)
Female	3,012	47.4 (45.5–49.2)	2,093	50.9 (48.8–52.9)	919	41.2 (38.2–44.1)
Race/ethnicity
Non-Hispanic White	3,543	59.6 (55.8–63.4)	2,339	60.6 (56.7–64.4)	1,204	57.9 (53.2–62.5)
Non-Hispanic Black	667	10.8 (9.0–12.9)	401	10.1 (8.4–12.1)	266	12.1 (9.7–14.9)
Hispanic	1,893	25.4 (22.2–28.9)	1,212	25.2 (21.9–28.8)	681	25.8 (21.8–30.1)
Non-Hispanic Other	277	4.2 (3.4–5.2)	176	4.1 (3.1–5.4)	101	4.3 (3.4–5.4)
Frequency of tobacco product use d
1–5 days	3,110	47.1 (45.0–49.2)	2,455	58.2 (55.8–60.6)	655	27.4 (24.8–30.0)
6–19 days	1,153	18.1 (17.0–19.2)	715	17.4 (16.2–18.8)	438	19.3 (17.3–21.4)
20 + days	2,150	34.8 (32.7–37.0)	963	24.3 (22.3–26.4)	1,187	53.3 (50.3–56.4)
Single product used
Cigarettes only	na	na	128	3.0 (2.3–3.8)	na	na
Cigars onlye	na	na	246	6.4 (5.1–8.1)	na	na
E-cigarettes only	na	na	3,385	80.5 (77.9–82.9)	na	na
HTPs only	na	na	86	2.1 (1.5–2.8)	na	na
Hookah only	na	na	138	3.3 (2.5–4.4)	na	na
Smokeless tobacco onlye	na	na	171	4.1 (3.1–5.3)	na	na
Flavored tobacco product use f
Yes	4,704	72.7 (70.9–74.5)	2,811	67.3 (65.1–69.5)	1,899	82.4 (80.0–84.5)
No	1,765	27.3 (25.5–29.1)	1,372	32.7 (30.5–34.9)	349	17.6 (15.5–20.0)
Age at first tobacco product use g
<11 years	993	15.7 (14.0–17.6)	365	9.1 (7.8–10.5)	628	27.3 (24.3–30.5)
≥11 years	5,351	84.3 (82.4–86.0)	3,708	90.9 (89.5–92.2)	1,643	72.7 (69.5–75.7)
Multiple (≥2) tobacco product use
Yes	2,286	35.9 (33.6–38.2)	na	na	na	na
No	4,185	64.1 (61.8–66.4)	na	na	na	na
Tobacco dependence symptom(s)
Reported one or more dependence symptom	1,917	32.0 (30.0–34.2)	846	22.5 (20.5–24.7)	1,071	49.0 (46.1–51.8)
Reported strong craving for tobacco product in past 30 days	1,711	28.3 (26.3–30.5)	748	19.5 (17.6–21.6)	963	44.0 (41.0–47.0)
Reported wanting to use tobacco within 30 min of waking	920	15.7 (14.2–17.3)	366	10.1 (8.8–11.6)	554	25.7 (22.8–28.8)

na = not applicable; 95 % CI = 95 % Confidence Interval.

^aDefined as use of cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, heated tobacco products (HTPs), hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables) on at least 1 day in the past 30 days.

^bDefined as use of only one of the following products on at least 1 day in the past 30 days: cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables).

^cDefined as use of two or more of the following products on at least 1 day in the past 30-days: cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables).

^dFrequency of tobacco product use was assessed among current users of the following products: e-cigarettes, cigarettes, cigars (regular cigars, cigarillos, or little cigars), smokeless tobacco (chewing tobacco, snuff, or dip), hookahs, and HTPs. Respondents were asked, “During the past 30 days, on how many days did you use [tobacco product]? If a respondent reported use of more than one of these products in the past 30 days, analyses of the number of days used was based on the product used most frequently. Frequency of use for snus, dissolvables, pipe tobacco, and bidis was not collected in the 2019 and 2020 surveys.

^eCigars include regular cigars, cigarillos, and little cigars. Smokeless tobacco includes chewing tobacco, snuff, dip, snus, and dissolvables. Pipe tobacco only (n = 19) and bidis only (n = 12) were included in the denominator.

^fDefined as use of one or more flavored tobacco products in the past 30 days. Flavored product use was assessed separately among current (past 30-day) users of cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookah, bidis, pipe tobacco, and smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables) and then recoded as a composite among current users of any tobacco product combined. In 2019, respondents were asked, “Which of the following tobacco products that you used in the past 30 days were flavored to taste like menthol (mint), alcohol (wine, cognac), candy, fruit, chocolate, or any other flavors?” Current users who selected from a list of tobacco products to indicate the flavored product(s) they had used were categorized as flavored users; those who indicated “I did not use any flavored tobacco products in the past 30 days” or those who did not select any product from the list were categorized as “No/Unknown” flavored users. In 2020, question structure was updated to assess flavored product use for each product separately. Current users of each respective product were asked, “Were any of the [tobacco product] that you used in the past 30 days flavored to taste like menthol, mint, clove or spice, alcohol (wine, cognac), candy, fruit, chocolate, or any other flavor?” If current users answered “Yes” to any of the flavored product questions, they were categorized as flavored users; if answered “No” or “Don’t know” for all the flavored product questions, they were categorized as “No/Unknown” flavored users. In 2019, flavored (menthol) cigarette use was ascertained if current cigarette smokers reported “Newport” or “Kool” as the usual cigarette brand or answered “Yes” to the question, “During the past 30 days, were the cigarettes that you usually smoked menthol?” In 2020, current cigarette smokers were categorized as flavored (menthol) cigarette smokers: if they responded “yes” to the question, “Menthol cigarettes are cigarettes that taste like mint. During the past 30 days, were the cigarettes that you usually smoked menthol?”; or if they indicated “Kool” or “Newport” as a brand they usually smoked in the past 30 days. Usual brand was determined based on responses to two questions: 1) “During the past 30 days, what brands of cigarettes did you smoke? (Select one or more)” and 2) “During the past 30 days, what brand of cigarettes did you usually smoke? (Choose only one answer).” If “Kool” or “Newport” was the only brand selected for the first question, or if multiple brands were selected in the first question and “Kool” or “Newport” was selected for the second question, “Kool” or “Newport” was considered the respondent’s usual brand. In 2019 and 2020, if current smokers answered “No” or “Not sure” to the menthol cigarette question or did not report “Newport” or “Kool” as the usual brand, they were categorized as “No/Unknown” menthol smokers.

^gAge at first tobacco product use was assessed for the following tobacco products: e-cigarettes, cigarettes, cigars (regular cigars, cigarillos, or little cigars), smokeless tobacco (chewing tobacco, snuff, or dip), and hookah. Respondents were asked, “How old were you when you first used/tried [tobacco product], even once or twice/even one or two puffs? If a respondent reported use of more than one of these products in the past 30 days, analyses of the age at first use was based on the earliest product used. Age at first use was not assessed for snus, dissolvables, pipe tobacco, bidis and HTPs in the 2019 and 2020 surveys.

Table 2 presents the prevalence and determinants of each tobacco dependence symptom among current tobacco users, single product users, and multiple product users. Generally, the prevalence of each dependence symptom was higher among multiple product users than single product users; for example, among those who reported using tobacco infrequently, a higher percentage of multiple product users reported having strong cravings than single product users (21.1 % vs 9.6 %). In multivariable models, moderate and frequent tobacco product use, flavored product use, initiating tobacco use < 11 years, and multiple product use were associated with higher odds of reporting strong cravings and desire to use tobacco within 30 min of waking, generally. Among all user groups (overall, single product, and multiple product), female students, those who reported moderate or frequent use, those who used flavored products, those who began using tobacco products < 11 years (except multiple product users), and multiple product users (overall only) had higher odds of reporting strong craving. Further, among all user groups (overall, single product, and multiple product), students who reported moderate or frequent product use, flavored product use (except multiple product users), those who initiated tobacco products < 11 (except multiple product users), and multiple product users (overall only) had higher odds of reporting a desire to use tobacco within 30 min of waking. Among single product users, those who smoked cigarettes or used flavored products also had higher odds of reporting strong cravings.

Table 2.

Characteristics associated with tobacco dependence symptoms during the past 30 days, 2019–2020 National Youth Tobacco Survey.

	Strong Cravings						Desire to use tobacco within 30 min of waking
	Current tobacco usersa Unweighted n = 1,711		Current single product usersb Unweighted n = 748		Current multiple product usersc Unweighted n = 963		Current tobacco usersa Unweighted n = 920		Current single product usersb Unweighted n = 366		Current multiple product usersc Unweighted n = 554
	% (95 % CI)	AOR (95 % CI)	% (95 % CI)	AOR (95 % CI)	% (95 % CI)	AOR (95 % CI)	% (95 % CI)	AOR (95 % CI)	% (95 % CI)	AOR (95 % CI)	% (95 % CI)	AOR (95 % CI)
OVERALL	28.3 (26.3–30.5)	na	19.5 (17.6–21.6)	na	44.0 (41.0–47.0)	na	15.7 (14.2–17.3)	na	10.1 (8.8–11.6)	na	25.7 (22.8–28.8)	na
School type
Middle school	24.6 (21.8–27.6)	1.2 (0.9–1.5)	15.3 (12.6–18.6)	1.1 (0.8–1.4)	40.4 (35.3–45.8)	1.2 (0.9–1.6)	9.6 (8.0–11.4)	0.8 (0.6–1.0)	4.4 (3.1–6.4)	0.6 (0.4–1.0)	18.4 (15.2–22.1)	0.9 (0.6–1.2)
High school	29.4 (27.0–31.9)	1.0	20.7 (18.5–23.2)	1.0	45.1 (41.8–48.4)	1.0	17.3 (15.6–19.2)	1.0	11.7 (10.1–13.4)	1.0	27.5 (24.2–31.1)	1.0
Sex
Male	27.1 (24.5–29.9)	1.0	17.4 (14.8–20.3)	1.0	41.5 (38.1–45.0)	1.0	17.1 (15.1–19.4)	1.0	10.5 (8.4–13.0)	1.0	27.0 (23.6–30.7)	1.0
Female	29.8 (27.3–32.4)	1.6 (1.4–1.9)	21.6 (19.3–24.2)	1.6 (1.3–1.9)	47.8 (43.2–52.4)	1.8 (1.4–2.3)	14.2 (12.3–16.2)	1.1 (0.9–1.4)	9.8 (8.4–11.5)	1.1 (0.8–1.4)	23.7 (19.9–28.1)	1.2 (0.9–1.5)
Race/ethnicity
Non-Hispanic White	32.5 (30.0–35.1)	1.0	22.9 (20.4–25.6)	1.0	50.4 (46.9–53.9)	1.0	18.8 (16.9–20.9)	1.0	12.9 (11.1–15.0)	1.0	29.8 (26.1–33.7)	1.0
Non-Hispanic Black	19.7 (16.1–23.9)	0.7 (0.5–1.0)	13.7 (9.7–19.1)	0.9 (0.5–1.5)	28.6 (21.4–37.2)	0.6 (0.4–0.9)	10.3 (7.3–14.4)	0.8 (0.5–1.4)	7.1 (4.2–11.7)	1.0 (0.4–2.1)	15.1 (9.4–23.3)	0.7 (0.4–1.3)
Hispanic	21.4 (18.7–24.3)	0.7 (0.6–0.9)	13.4 (11.1–16.2)	0.8 (0.6–1.0)	35.1 (30.1–40.4)	0.6 (0.5–0.8)	10.4 (8.5–12.7)	0.7 (0.6–0.9)	5.2 (3.9–7.0)	0.7 (0.4–1.0)	19.4 (15.3–24.4)	0.7 (0.5–1.0)
Non-Hispanic Other	32.7 (25.9–40.2)	1.1 (0.8–1.6)	22.6 (15.2–32.3)	1.1 (0.6–2.0)	50.0 (38.3–61.6)	1.0 (0.6–1.7)	16.3 (11.6–22.4)	0.9 (0.6–1.4)	--d	0.5 (0.3–1.0)	32.6 (22.3–44.8)	1.2 (0.6–2.2)
Frequency of tobacco product use e
1–5 days	12.0 (10.5–13.7)	1.0	9.6 (8.1–11.3)	1.0	21.1 (17.4–25.3)	1.0	2.7 (1.9–3.7)	1.0	2.4 (1.6–3.7)	1.0	3.6 (2.3–5.8)	1.0
6–19 days	26.3 (22.8–30.2)	2.1 (1.7–2.7)	21.3 (17.6–25.6)	2.5 (1.9–3.3)	34.1 (28.5–40.3)	1.7 (1.2–2.5)	9.9 (7.9–12.3)	3.1 (2.0–4.8)	5.7 (4.1–7.9)	2.2 (1.2–3.8)	16.4 (12.4–21.5)	4.2 (2.4–7.3)
20 + days	51.7 (48.5–54.8)	5.9 (4.9–7.2)	41.8 (37.7–46.0)	6.8 (5.4–8.6)	59.6 (55.7–63.3)	5.2 (3.8–7.2)	36.5 (33.4–39.8)	15.1 (10.6–21.5)	31.7 (27.9–35.7)	16.1 (10.2–25.4)	40.5 (36.0–45.0)	14.9 (8.6–25.8)
Single product used
Cigarettes only	na	na	34.2 (23.7–46.4)	4.3 (2.5–7.4)	na	na	na	na	--d	1.9 (0.7–4.8)	na	na
Cigars onlyf	na	na	14.8 (10.0–21.5)	1.4 (0.7–2.7)	na	na	na	na	--d	1.6 (0.7–3.6)	na	na
E-cigarettes only	na	na	19.4 (17.5–21.5)	1.0	na	na	na	na	10.4 (9.1–11.9)	1.0	na	na
HTPs only	na	na	--d	1.6 (0.4–6.5)	na	na	na	na	--d	0.3 (0.0–2.4)	na	na
Hookah only	na	na	15.9 (9.3–25.8)	1.6 (0.8–3.1)	na	na	na	na	--d	1.9 (0.5–6.7)	na	na
Smokeless tobacco onlyf	na	na	24.9 (17.8–33.6)	1.7 (1.0–2.9)	na	na	na	na	15.8 (9.0–26.4)	1.4 (0.7–2.8)	na	na
Flavored tobacco product use g
Yes	33.3 (31.0–35.8)	1.7 (1.4–2.2)	23.6 (21.4–26.0)	2.2 (1.7–2.9)	47.5 (44.3–50.7)	1.7 (1.2–2.4)	18.8 (17.0–20.7)	1.6 (1.2–2.1)	12.6 (11.0–14.4)	2.2 (1.4–3.5)	27.9 (24.8–31.1)	1.3 (0.8–2.1)
No	14.7 (12.6–17.0)	1.0	10.9 (8.7–13.6)	1.0	27.2 (21.8–33.4)	1.0	7.2 (5.7–9.1)	1.0	4.8 (3.2–7.1)	1.0	15.2 (10.5–21.5)	1.0
Age at first tobacco product use h
<11 years	42.8 (38.9–46.9)	1.4 (1.1–1.7)	27.9 (21.6–35.1)	1.8 (1.2–2.7)	51.4 (46.8–56.0)	1.2 (0.9–1.6)	26.3 (22.8–30.2)	1.4 (1.1 −1.8)	13.1 (9.1–18.4)	1.6 (1.1–2.3)	34.0 (29.3–39.1)	1.3 (1.0–1.8)
≥11 years	26.0 (24.1–28.1)	1.0	19.0 (17.2–21.1)	1.0	41.4 (38.1–44.6)	1.0	13.9 (12.5–15.4)	1.0	9.9 (8.6–11.4)	1.0	22.6 (19.7–25.8)	1.0
Multiple (≥2) tobacco product use
Yes	44.0 (41.0–47.0)	2.1 (1.7–2.5)	na	na	na	na	25.7 (22.8–28.8)	1.5 (1.2–1.9)	na	na	na	na
No	19.5 (17.6–21.6)	1.0	na	na	na	na	10.1 (8.8–11.6)	1.0	na	na	na	na

na = Not applicable; 95 % CI = 95 % Confidence Interval; AOR = adjusted odds ratio.

Note: All regression estimates were mutually adjusted for the following covariates: sex, school type, race/ethnicity, age at first use, frequency of use, and flavored product use. Among current tobacco users, multiple product use was included as a covariate. Among single product users, the type of tobacco product used was included as a covariate..

^aDefined as use of cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, heated tobacco products (HTPs), hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables) on at least 1 day in the past 30 days.

^bDefined as use of only one of the following products on at least 1 day in the past 30 days: cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables).

^cDefined as use of two or more of the following products on at least 1 day in the past 30-days: cigarettes, cigars (Regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookah, bidis, pipe tobacco, or smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables).

^dDashes indicate data are statistically unreliable and suppressed because the unweighted denominator was < 50 or the relative standard error was > 30 %.

^eFrequency of tobacco product use was assessed among current users of the following products: e-cigarettes, cigarettes, cigars (cigars, cigarillos, or little cigars), smokeless tobacco (chewing tobacco, snuff, or dip), hookahs, and HTPs. Respondents were asked, “During the past 30 days, on how many days did you use [tobacco product]? If a respondent reported use of more than one of these products in the past 30 days, analyses of the number of days used was based on the product used most frequently. Frequency of use for snus, dissolvables, pipe tobacco, and bidis was not collected in the 2019 and 2020 surveys.

^fCigars include regular cigars, cigarillos, and little cigars. Smokeless tobacco includes chewing tobacco, snuff, dip, snus, and dissolvables. Pipe tobacco only (n = 19) and bidis only (n = 12) were included in the denominator.

^gDefined as use of one or more flavored tobacco products in the past 30 days. Flavored product use was assessed separately among current (past 30-day) users of cigarettes, cigars (regular cigars, cigarillos, or little cigars), e-cigarettes, HTPs, hookahs, bidis, pipe tobacco, and smokeless tobacco (chewing tobacco, snuff, dip, snus, or dissolvables) and then recoded as a composite among current users of any tobacco product combined. In 2019, respondents were asked, “Which of the following tobacco products that you used in the past 30 days were flavored to taste like menthol (mint), alcohol (wine, cognac), candy, fruit, chocolate, or any other flavors?” Current users who selected from a list of tobacco products to indicate the flavored product(s) they had used were categorized as flavored users; those who indicated “I did not use any flavored tobacco products in the past 30 days” or those who did not select any product from the list were categorized as “No/Unknown” flavored users. In 2020, question structure was updated to assess flavored product use for each product separately. Current users of each respective product were asked, “Were any of the [tobacco product] that you used in the past 30 days flavored to taste like menthol, mint, clove or spice, alcohol (wine, cognac), candy, fruit, chocolate, or any other flavor?” If current users answered “Yes” to any of the flavored product questions, they were categorized as flavored users; if answered “No” or “Don’t know” for all the flavored product questions, they were categorized as “No/Unknown” flavored users. In 2019, flavored (menthol) cigarette use was ascertained if current cigarette smokers reported “Newport” or “Kool” as the usual cigarette brand or answered “Yes” to the question, “During the past 30 days, were the cigarettes that you usually smoked menthol?” In 2020, current cigarette smokers were categorized as flavored (menthol) cigarette smokers: if they responded “yes” to the question, “Menthol cigarettes are cigarettes that taste like mint. During the past 30 days, were the cigarettes that you usually smoked menthol?”; or if they indicated “Kool” or “Newport” as a brand they usually smoked in the past 30 days. Usual brand was determined based on responses to two questions: 1) “During the past 30 days, what brands of cigarettes did you smoke? (Select one or more)” and 2) “During the past 30 days, what brand of cigarettes did you usually smoke? (Choose only one answer).” If “Kool” or “Newport” was the only brand selected for the first question, or if multiple brands were selected in the first question and “Kool” or “Newport” was selected for the second question, “Kool” or “Newport” was considered the respondent’s usual brand. In 2019 and 2020, if current smokers answered “No” or “Not sure” to the menthol cigarette question or did not report “Newport” or “Kool” as the usual brand, they were categorized as “No/Unknown” menthol smokers.

^hAge at first tobacco product use was assessed for the following tobacco products: e-cigarettes, cigarettes, cigars (cigars, cigarillos, or little cigars), smokeless tobacco (chewing tobacco, snuff, or dip), and hookahs. Respondents were asked, “How old were you when you first used/tried [tobacco product], even once or twice/even one or two puffs? If a respondent reported use of more than one of these products in the past 30 days, analyses of the age at first use was based on the earliest product used. Age at first use was not assessed for snus, dissolvables, pipe tobacco, bidis, and HTPs in the 2019 and 2020 surveys.

4. Discussion

This study found similar and complementary findings to those reported in Apelberg et al. (2014). However, Apelberg and colleagues estimated that 62.4 % of current tobacco users were multiple product users, compared to 35.9 % in this study, largely due to the majority of youth e-cigarette users being single product users. The current study found that reporting both dependence symptoms was associated with multiple tobacco product use and increased frequency of use. Further-more, for both single and multiple product users, those who used flavored tobacco products, female users, and those who began using tobacco products < 11 years (single product users only) had higher odds of reporting strong cravings. Consistent with findings from Gomez and colleagues (Gomez et al., 2020), a higher percentage of multiple product users reported symptoms of dependence than single product users, which could have effects on overall nicotine exposure; (Chen et al., 2014; Choi et al., 2017; Strong et al., 2017) however, biological confirmation or total nicotine exposure are not captured in NYTS due to the self-reported nature of the survey.

Increased dependence symptoms in youth could indicate reinforcing addiction leading to sustained tobacco use in young adulthood, and future health problems (National Center for Chronic Disease Prevention, 2012). This study found a positive, significant relationship between the number of days tobacco products used and dependence symptoms; even infrequent users reported dependence symptoms, especially those who were multiple product users. The strength of the association between frequency of use and the dependence symptoms increased as frequency of use increased, consistent with findings from Apelberg et al. (2014) Use of flavored tobacco products was associated with reporting strong cravings for single and multiple product users. These results are consistent with other published studies showing that flavors are likely to impact patterns of tobacco product use (Barrington-Trimis et al., 2018; Chen et al., 2017; Dai & Hao, 2016; Leventhal et al., 2019).

4.1. Limitations

This study is subject to multiple limitations. First, these results may not be generalizable to all U.S. adolescents, as the survey is not conducted among youth who are home schooled, have dropped out of school, or are in detention centers. Second, this study only used two questions to assess tobacco dependence; additional measurement items are likely needed to assess the heterogeneity among tobacco users classified as dependent, while it is also possible that there is misclassification among tobacco users classified as non-dependent, particularly among infrequent users.. Third, the inclusion of snus and dissolvables in the smokeless tobacco definition could lead to misclassification if students selected more than one smokeless tobacco product but were classified as single product users. Fourth, given the cross-sectional nature of this survey, a temporal relationship between tobacco use and dependence could not be evaluated. Lastly, these data are self-reported and subject to recall or reporting bias.

5. Conclusions

Consistent with prior research, results from this study show that a considerable amount of U.S. adolescent tobacco users, even infrequent users, report symptoms of dependence. Among the estimated 5.36 million adolescents who reported current tobacco use, about one-third reported at least one symptom of tobacco dependence. Reporting of dependence symptoms was generally associated with multiple tobacco product use, higher frequency of use, initiating tobacco use < 11 years, and use of flavored products. These findings highlight the continued importance of youth tobacco initiation reduction strategies and can inform policy development.

Submission declaration and verification: The work described herein has not been published previously. Further, it is not under consideration for publication elsewhere, is approved by all authors, and if accepted, will not be published elsewhere in the same form, in English or in any other language, including electronically without the written consent of the copyright-holder.