Adolescent Risk and Protective Factors for the Use of Electronic Cigarettes

Abstract

The use of electronic cigarettes (e-cigarettes) by adolescents is a serious public health concern. The major aim of the current study is to conduct a comprehensive examination of adolescent e-cigarette use in relation to risk and protective factors for a school-based sample. The present study is based on a secondary data analysis of the 2017 Prevention Needs Assessment (PNA) Survey that is administered every 2 years in the state of Utah to a large sample of students (n = 54,853) in Grades 6, 8, 10, and 12 from more than 400 schools. The PNA Survey measures substance use, mental health symptoms, and antisocial behavior as well as their associated risk and protective factors. Almost 9% of adolescents in this study reported using e-cigarettes in the past 30 days. Adolescents who reported infrequent (1–5 days) and frequent (6+ days) use of e-cigarettes also indicated lower levels of protection (e.g., perceived harm) and higher levels of risk (e.g., favorable attitudes) compared with students who did not report using e-cigarettes in the past 30 days. The frequency of adolescent e-cigarette use can distinguish between risk and protective factors. Findings suggest that the risk and protective factors relevant for adolescent alcohol, cannabis, and tobacco use extend to the use of the e-cigarettes. Certain risk factors (e.g., favorable attitudes toward substance use) and protective factors (e.g., perceived risk for use of e-cigarettes) hold promise for preventive interventions in addressing this public health concern.

Keywords

electronic cigarettes adolescents school-based survey risk factors protective factors

The use of electronic cigarettes (e-cigarettes) by adolescents is a serious public health concern, and rates of use are on the rise in the United States. Between 2011 and 2018, there has been a marked increase in the number of middle and high school students using e-cigarettes with a corresponding decline in the use of combustible cigarettes (Gentzke et al., 2019). Recent estimates suggest that more than 27% (or >4 million) of high school students and almost 11% (or 1.2 million) of middle school students have used e-cigarettes at least once in the past 30 days (Cullen et al., 2019). The rise of youth e-cigarette use causes concern for many reasons. First, e-cigarettes frequently contain nicotine even when youth believe the liquids only contain flavoring due to the mislabeling of packaging (Goniewicz et al., 2015; Omaiye et al., 2017; Raymond et al., 2018). Second, the adolescent brain is primed for exposure to nicotine that can have deleterious effects on subsequent development and enhance the rewarding experience of other substances such as alcohol (Yuan et al., 2015). Third, the vaporized aerosol produced by e-cigarettes can contain harmful chemicals and heavy metals that may lead to cardiovascular or other health problems (Bhatta & Glantz, 2019). Fourth, there is evidence to suggest that the use of e-cigarettes increases the risk of subsequent combustible cigarette use (Chaffee et al., 2018). Finally, the hospitalizations and deaths seen in the popular media linked to e-cigarette use has caused alarm that certain chemicals found in these products (e.g., vitamin E acetate) may contribute to serious lung injury (e.g., electronic cigarette or vaping product use–associated lung injury [EVALI]), respiratory disease, and death (Bhatta & Glantz, 2019; Blount et al., 2019; Schier et al., 2019). Despite these associated risks, the use of e-cigarettes by youth continues to rise (Gentzke et al., 2019).

Recent prevalence rates indicate that 11% and 27% of middle and high school students have used e-cigarettes on at least 1 or more days in the past month, respectively (Cullen et al., 2019). Therefore, an important question to ask is “Why do some students use e-cigarettes and others do not?” Theoretically, this question can be addressed via the Social Development Model (SDM; Catalano & Hawkins, 1996; Hawkins & Weis, 1985) that describes prosocial and antisocial pathways for the development of substance use in youth. The SDM posits that youths’ exposure to risk and protective factors influence behavior in predictive ways and occur within the five youth socializing domains of individual, peers, family, school, and community (Hawkins et al., 1986, 1992). A risk factor is a characteristic of an individual or environment that increases the chances of using a substance whereas a protective factor is something that decreases such chances (Hawkins et al., 1992). Risk and protective factors may have complex relationships that are not simply the opposites of the other. For example, some risk factors are difficult to eliminate (e.g., availability of substances), whereas a protective factor (e.g., perceived risk of using substances) can buffer the effect of a particular risk factor (Catalano & Hawkins, 1996; Hawkins et al., 1992). Risk and protective factors are not limited to a specific substance (e.g., alcohol); rather, they may be applicable to multiple substances. From the perspective of the SDM, risk and protective factors are categorized in five socializing domains for youth. For example, an adolescent’s favorable attitude toward substance use is an individual risk factor (Guo et al., 2001), whereas recognition from teachers for displaying prosocial behavior is a school-based protective factor (Jackson et al., 2005). The applicability of risk and protective factors for adolescent substance use such as alcohol, cannabis, and tobacco have been examined in prior research (see Guo et al., 2001; Meier et al., 2019; Scal et al., 2003), whereas corresponding analysis for e-cigarettes is limited.

To date, there have only been a handful of studies examining the risk and protective factors for adolescent e-cigarette use. Wills and colleagues (2015) surveyed a sample of high school students (N = 1,941; M age = 14.6 years) in Hawaii. The researchers found that students who used combustible cigarettes and e-cigarettes (i.e., dual use) or combustible cigarettes only reported higher risk factors (e.g., parent–child conflict) and lower protective factors (e.g., parental support) compared with youth who reported using e-cigarettes only or reported not using either substance. In addition, the youth who reported dual use as well as e-cigarettes only perceived e-cigarettes as healthier compared with youth who did not use either substance. A second study by Dunbar and colleagues (2017) conducted a secondary analysis with a sample of adolescents (N = 2,488, M age = 17.3 years) from greater Los Angeles, California, that were originally part of a larger evaluation study beginning when the students were in middle school. Youth were asked about their use of substances that included e-cigarettes and combustible cigarettes as well as risk and protective health-related behaviors (e.g., physical activity, mental health, diet, sleep). The researchers compared four groups of adolescents based on their use of e-cigarettes only, dual use (i.e., e-cigarette/combustible cigarette), combustible cigarette use only, or nonuse of either substance. In general, the researchers found that all three cigarette-related user groups demonstrated higher risk and less protection on health-related behaviors compared with those who did not use either substance. One exception was for youth who used e-cigarettes only; these students endorsed fewer mental health symptoms and lower use of other substances (e.g., alcohol, cannabis) compared with either dual or combustible cigarette users. A third study conducted by Kwon and colleagues (2018) was based on secondary data analysis of a sample of adolescents (N = 9,853, between 12 and 17 years old) derived from the Population Assessment of Tobacco and Health (PATH) study. The PATH study is a national household survey sponsored by the National Institute on Drug Abuse and the U.S. Food and Drug Administration Center for Tobacco Products. The goal of the Kwon et al. study was to examine the susceptibility for using e-cigarettes in a subsample of adolescents who reported not using either combustible or e-cigarettes (n = 2,410). They found that reporting higher levels of psychological problems (i.e., depressive symptoms) and rebelliousness (i.e., rule breaking, impulsivity, and sensation seeking) was predictive of higher susceptibility to e-cigarette use for youth in the sample. Furthermore, the researchers indicated that using other substances (e.g., alcohol, cannabis) was a risk factor, whereas the perception of e-cigarettes as harmful was a protective factor for the susceptibility of adolescent e-cigarette use. As reviewed above, a small number of studies exist on the applicability of risk and protective factors for adolescent e-cigarette use, whereas a comprehensive examination of risk and protective factors across the five socializing domains has yet to be conducted.

The purpose of the current study is to conduct a comprehensive examination of adolescent e-cigarette use in relation to risk and protective factors for a school-based sample. To address this study purpose, we proposed three research questions:

Research Question 1: Does the frequency of adolescent e-cigarette use predict differences in lifetime substance use?

Research Question 2: Does the frequency of adolescent e-cigarette use predict differences in early initiation of other substances?

Research Question 3: Does the frequency of adolescent e-cigarette use predict differences in levels of risk and protective factors?

Increasing our understanding of adolescent e-cigarette use in relation to risk and protective factors is important for determining areas that have the most potential for preventive interventions.

Method

The present study is based on a secondary data analysis of the 2017 Prevention Needs Assessment (PNA) Survey that is administered every 2 years by the Utah Department of Human Services (UDHS) Division of Substance Use and Mental Health to a large state sample of students in 6th, 8th, 10th, and 12th grades (Bach Harrison, 2017; UDHS, 2019). The PNA is based on the Communities that Care (CTC) Youth Survey (see Arthur et al., 2002) and is designed to measure substance use, mental health symptoms, and antisocial behavior as well as their associated risk and protective factors. The psychometric properties of the CTC Youth Survey and item scales for risk and protective factors have been psychometrically validated across a number of studies (see Arthur et al., 2002, 2007; Briney et al., 2012; Feinberg et al., 2007; Glaser et al., 2005; Kuttler et al., 2015). The PNA was administered using a complex sampling framework; in particular, strata were at the school district level, clustering at the school level, and weights were calculated for the sample to approximate the population characteristics. In the spring 2017, the PNA was administered to 54,853 students in more than 400 public or charter schools across the state of Utah. This sample represented 29.7% of the population of students in Grades 6, 8, 10, and 12. The survey was administered to students in paper (62.5%) or online (37.4%) formats. Active parental consent was used to recruit students to participate in the survey. The survey did not ask students for identifying information, and their responses remained anonymous. A total of 1,144 students were excluded from the final sample due to reporting being “Not Honest At All” during survey completion (n = 338), using a fake substance (i.e., phenoxydine; n = 693), using an unrealistically high level of substances (n = 284), a past-month substance use rate that was higher than their lifetime substance use rate (n = 223), or an age inconsistent with their grade or school (n = 167). In addition, 3,369 students were excluded because they reported being in 7th, 9th, or 11th grade, and 103 were excluded because they did not mark or marked multiple grade levels, resulting in a total sample of 50,237 students. Two versions of the PNA, Versions A (n = 25,952) and B (n = 24,285), were administered randomly during a standard class period. Most items were included on both versions of the survey, whereas some items were only included on version A or B. The sample weights were calculated for the complete survey as well as for the two versions of the survey (i.e., A or B).

Past 30-Day Use of E-Cigarettes

The frequency of past 30-day e-cigarette use was determined through participant responses to a single multiple-choice item:

During the past 30 days, on how many days did you use electronic cigarettes or e-cigarettes (0 days, 1 or 2 days, 3 to 5 days, 6 to 9 days, 10 to 19 days, 20 to 29 days, all 30 days)?

Based on their response, students in the sample were classified into one of the three groups: those reporting 0 days in a non-e-cigarette use group (n = 44,583), 1 to 5 days in an infrequent e-cigarette use group (n = 2,371), and 6 or more days in a frequent e-cigarette use group (n = 1,398). The three groups (i.e., 0 days, 1–5 days, and 6+ days) were based on a visual inspection of the distribution to determine the most common break points.

Early Initiation and Lifetime Use of Cigarettes, Alcohol, and Cannabis

Early initiation and lifetime use of cigarettes, alcohol, and cannabis were determined by participant responses to three multiple-choice items pertaining to the age of initiation for each substance. Specifically, the question asked, “How old were you when you first used [cigarettes, alcohol, or marijuana] (Never, 10 or younger, 11, 12, 13, 14, 15, 16, and 17 or older)?” A dichotomous variable was created, those who responded “Never” were classified as not having used the substance in their lifetime and those who responded with anything other than “Never” were classified as having used the substance in their lifetime. Early initiation was defined as using a substance before the age of 14 years. A second dichotomous variable was created; those who responded “10 or younger,” “11,” “12,” or “13” were classified as participating in early initiation of substance use and those who responded anything else were classified as not participating in early initiation of substance use. Approximately half of the students who indicated lifetime cigarette (54.5%) or alcohol (48.8%) use reported use prior to the age of 14 years, whereas about a third of students who indicated lifetime cannabis use (36.3%) reported use before the age of 14 years.

Risk and Protective Factors

The individual risk and protective factor items on the PNA Survey were based on the CTC (Arthur et al., 2002) Youth Survey. These items were developed to measure the risk and protective factors for substance use and related problem behaviors and have been validated in multiple studies (see Arthur et al., 2002; Kim et al., 2015; Monahan et al., 2014). The individual items on the PNA were combined to produce 32 risk and protective factor subscales across the five domains of individual, peer, family, school, and community. Table 1 contains the risk and protective factor subscales, item descriptions, and estimates of reliability across each of the five domains. Of the 32 risk and protective factors, one risk factor was coded as dichotomous (i.e., family history of alcohol and/or drug use = yes or no) and the remaining 31 items were coded as continuous (i.e., 4- or 5-point scale) depending on the number of item response options contained in the survey.

Table 1.

Risk and Protective Factors, Reliability, Item Totals, Descriptions, and Response Options.

Risk or protective factor	Reliability (α)	No. of items	Item description(s)	Response options
Individual factors
Perceived risk of harm from marijuana	.883	2	How much do you think people risk harming themselves (physically or in other ways) if they: • try marijuana once or twice? • smoke marijuana regularly?	No risk, Slight risk, Moderate risk, Great risk
Perceived risk of harm from alcohol	NA	1	How much do you think people risk harming themselves (physically or in other ways) if they take one or two drinks of an alcoholic beverage (beer, wine, liquor) nearly every day.	No risk, Slight risk, Moderate risk, Great risk
Perceived risk of harm from vaping	NA	1	How much do you think people risk harming themselves (physically or in other ways) if they use vape products such as e-cigarettes, vape pens, or mods?	No risk, Slight risk, Moderate risk, Great risk
Rebelliousness	.784	3	I do the opposite of what people tell me, just to get them mad. I like to see how much I can get away with. I ignore the rules that get in my way.	Very false, Somewhat false, Somewhat true, Very true
Favorable attitudes toward antisocial behavior	.758	5	How wrong do you think it is for someone your age to: • take a handgun to school? • steal anything worth more than US$5? • pick a fight with someone? • attack someone with the idea of seriously hurting them? • stay away from school all day when their parents think they are at school?	Very wrong, Wrong, A little bit wrong, Not wrong at all
Favorable attitudes toward substance use	.831	4	How wrong do you think it is for someone your age to: • drink beer, wine, or hard liquor (e.g., vodka, whiskey, or gin) regularly? • smoke cigarettes? • smoke marijuana? • use LSD, cocaine, amphetamines, or another illegal drug?	Very wrong, Wrong, A little bit wrong, Not wrong at all
Belief in a moral order	.685	4	I think sometimes it is okay to cheat at school.^a I think it is okay to take something without asking if you can get away with it.^a It is all right to beat up people if they start the fight.^a It is important to be honest with your parents, even if they become upset or you get punished.	NO!, no, yes, YES!
Prosocial involvement	.674	3	How many times in the past year (12 months) have you: • participated in clubs, organizations, and activities at school? • done extra work on your own for school? • volunteered to do community service?	Never, 1 or 2 times, 3–5, 6–9, 10–19, 20–29, 30–39, 40+^b
Peer factors
Peer substance use	.810	4	Think of your four best friends (the friends you feel closest to). In the past year (12 months), how many of your best friends have: • smoked cigarettes? • tried beer, wine, or hard liquor (e.g., vodka, whiskey, or gin) when their parents did not know about it? • used marijuana? • used LSD, cocaine, amphetamines, or other illegal drugs?	0, 1, 2, 3, 4
Peer favorable attitudes toward drug use	.867	4	How wrong do your friends feel it would be for you to: • have one or two drinks of an alcoholic beverage nearly every day? • smoke tobacco? • smoke marijuana? • use prescription drugs not prescribed to you?	Very wrong, Wrong, A little bit wrong, Not wrong at all
Being perceived as “cool” for cigarette use	NA	1	What are the chances you would be seen as cool if you smoked cigarettes?	No or very little chance, Little chance, Some chance, Pretty good chance, Very good chance
Being perceived as “cool” for alcohol use	NA	1	What are the chances you would be seen as cool if you began drinking alcoholic beverages regularly, that is, at least once or twice a month?	No or very little chance, Little chance, Some chance, Pretty good chance, Very good chance
Being perceived as “cool” for cannabis use	NA	1	What are the chances you would be seen as cool if you smoked marijuana?	No or very little chance, Little chance, Some chance, Pretty good chance, Very good chance
Being perceived as “cool” for substance use	.896	3	What are the chances you would be seen as cool if you: • smoked cigarettes? • began drinking alcoholic beverages regularly, that is, at least once or twice a month? • smoked marijuana?	No or very little chance, Little chance, Some chance, Pretty good chance, Very good chance
Interactions with antisocial peers	.671	6	Think of your four best friends (the friends you feel closest to). In the past year (12 months), how many of your best friends have: • been suspended from school? • carried a handgun? • sold illegal drugs? • stolen or tried to steal a motor vehicle such as a car or motorcycle? • been arrested? • dropped out of school?	None, 1, 2, 3, 4
Being perceived as “cool” for prosocial involvement	.806	3	What are the chances you would be seen as cool if you: • worked hard in school? • defended someone who was being verbally abused at school? • regularly volunteered to do community service?	No or very little chance, Little chance, Some chance, Pretty good chance, Very good chance
Interaction with prosocial peers	.745	5	Think of your four best friends (the friends you feel closest to). In the past year (12 months), how many of your best friends have: • participated in clubs, organizations, and activities at school? • made a commitment to stay drug-free? • tried to do well in school? • liked school? • regularly attended religious services?	0, 1, 2, 3, 4
Family factors
Family history of alcohol and/or drug use	.753	4	Has anyone in your family ever had severe alcohol or drug problems? Have any of your brothers or sisters ever: • drunk beer, wine, or hard liquor (e.g., vodka, whiskey, or gin)? • smoked marijuana? • smoked cigarettes?	No, Yes No, Yes, I do not have any brothers or sisters^c
Family conflict	.817	3	People in my family often insult or yell at each other. We argue about the same things in my family over and over. People in my family have serious arguments.	NO!, no, yes, YES!
Poor family management	.889	8	My parents ask if I have gotten my homework done. Would your parents know if you did not come home on time? The rules in my family are clear. When I am not at home, one of my parents knows where I am and who I am with. If you drank some beer or wine or liquor (e.g., vodka, whiskey, or gin) without your parents’ permission, would you be caught by your parents? My family has clear rules about alcohol and drug use. If you carried a handgun without your parents’ permission, would you be caught by your parents? If you skipped school, would you be caught by your parents?	NO!, no, yes, YES!
Parental favorable attitudes toward drug use	.740	3	How wrong do your parents feel it would be for YOU to: • drunk beer, wine, or hard liquor (e.g., vodka, whiskey, or gin)? • smoked marijuana? • smoked cigarettes?	Very wrong, Wrong, A little bit wrong, Not wrong at all
Family attachment	.811	4	Do you feel very close to your mother? Do you share your thoughts and feeling with your mother? Do you feel very close to your father? Do you share your thoughts and feeling with your father?	NO!, no, yes, YES!
Prosocial opportunities	.816	3	My parents ask me what I think before most family decisions affecting me are made. If I had a personal problem, I could ask my mom or dad for help. My parents give me lots of chances to do fun things with them.	NO!, no, yes, YES!
Prosocial rewards	.778	4	Do you enjoy spending time with your mother? Do you enjoy spending time with your father? My parents notice when I am doing a good job, and let me know about it. How often do your parents tell you they are proud of you for something you have done?	NO!, no, yes, YES! Never or almost never, Sometimes, Often, All the time
School factors
Academic failure	.693	2	Are your school grades better than the grades of most students in your class? Putting them all together, what were your grades like last year?	NO!, no, yes, YES! Mostly F’s, Mostly D’s, Mostly C’s, Mostly B’s, Mostly A’s^d
Low commitment to school	.796	7	Now, thinking back over the past year in school, how often did you: • enjoy being in school?^a • hate being in school? • try to do your best work in school?^a How often do you feel that the school work you are assigned is meaningful and important?^a How important do you think the things you are learning in school are going to be for your later life? How interesting are most of your courses to you? During the LAST 4 WEEKS, how many whole days of school have you missed because you skipped or “cut?”	Never, Seldom, Sometimes, Often, Almost always Almost always, Often, Sometimes, Seldom, Never Very important, Quite important, Fairly important, Slightly important, Not at all important Very interesting and stimulating, Quite interesting, Fairly interesting, Slightly interesting, Not at all interesting None, 1, 2, 3, 4–5, 6–10, 11 or more days^e
Prosocial opportunities	.608	5	In my school, students have lots of chances to help decide things like class activities and rules. Teachers ask me to work on special classroom projects. There are lots of chances for students in my school to get involved in sports, clubs, and other school activities outside of class. There are lots of chances for students in my school to talk with a teacher one-on-one. I have lots of chances to be part of class discussions or activities.	NO!, no, yes, YES!
Prosocial rewards	.742	4	My teachers notice when I am doing a good job and lets me know about it. I feel safe at my school. The school lets my parents know when I have done something well. My teachers praise me when I work hard in school.	NO!, no, yes, YES!
Community factors
Laws and norms favorable to drug use	.806	6	How wrong would most adults in your neighborhood think it was for kids your age: • to use marijuana. • to drink alcohol. • to smoke cigarettes. If a kid smoked marijuana in your neighborhood, would he or she be caught by the police?^a If a kid drank some beer, wine, or hard liquor (e.g., vodka, whiskey, or gin) in your neighborhood, would he or she be caught by the police?^a If a kid carried a handgun in your neighborhood, would he or she be caught by the police?^a	Very wrong, Wrong, A little bit wrong, Not wrong at all NO!, no, yes, YES!
Perceived availability of drugs	.890	4	If you wanted to get some cigarettes, how easy would it be for you to get some? If you wanted to get some beer, wine, or hard liquor (e.g., vodka, whiskey, or gin), how easy would it be for you to get some? If you wanted to get a drug like cocaine, LSD, or amphetamines, how easy would it be for you to get some? If you wanted to get some marijuana, how easy would it be for you to get some?	Very hard, Sort of hard, Sort of easy, Very easy
Low neighborhood attachment	.847	3	If I had to move, I would miss the neighborhood I now live in.^a I like my neighborhood.^a I would like to get out of my neighborhood.	NO!, no, yes, YES!
Rewards for prosocial involvement	.897	3	My neighbors notice when I am doing a good job and let me know about it. There are people in my neighborhood who are proud of me when I do something well. There are people in my neighborhood who encourage me to do my best.	NO!, no, yes, YES!

Note. LSD = lysergic acid diethylamide.

Item was reverse coded to reflect direction of risk or protective factor. ^b Reponses of “Never” and “1 or 2 times” were coded as 1, responses of “3–5” and “6–9” were coded as 2, responses of “10–19” and “20–29” were coded as 3, responses of “30–39” and “40+” were coded as 4, resulting in an item with four levels. ^c Responses of “I don’t have any brothers or sisters” and “No” were both coded as 0, resulting in a dichotomous item. ^d Responses of “Mostly F’s” and “Mostly D’s” were both coded as 1, resulting in an item with four levels. ^e Reponses of “None” were coded as 1, responses of “1” and “2” were coded as 2, responses of “3” were coded as 3, responses of “4–5” were coded as 4, responses of “6–10” and “11 or more days” were coded as 5, resulting in an item with five levels.

Analytical Plan

The independent variable for the primary research question was past 30-day use of e-cigarettes and constructed by dividing the sample into three subgroups based on the reported level of e-cigarette use in the past 30 days (i.e., no use, infrequent use, and frequent use). The three major areas of dependent variables (i.e., lifetime substance use, early initiation of substance use, or risk/protective factors) were coded either dichotomously or continuously based on the details provided above. For example, dependent variables such as lifetime substance use and early initiation were coded as dichotomous (i.e., lifetime use of a substance = “yes” or “no” or early initiation = “before” or “after” age 14 years), whereas almost all the risk and protective factors were coded as continuous based on a 4- or 5-item response scale. The data collected for the PNA was based on a complex survey design, and these elements were integrated into the analysis. Specifically, a complex survey analysis was utilized that included strata (district), cluster (school), and calculated sample weights. Sample weights for the overall sample and both versions of the survey (i.e., A and B) were calculated for the PNA by the original survey administrators using iteration proportional fitting using the variables grade, race/ethnicity, district, gender by district, race/ethnicity by district, and grade by district.

All analysis was conducted using Statistical Package for the Social Sciences (2019) v.26 Complex Analysis that can account for the strata, cluster, and weighted variables to produce accurate unweighted (sample) and weighted (population) estimates for the data. A complex samples analysis for logistic regression was used when the dependent variable was dichotomous (e.g., lifetime substance use, early initiation of substance use), whereas a complex samples linear regression was used when the outcome variable was continuous (i.e., majority of risk and protective factors). A total of 7 logistic regressions and 31 linear regressions were conducted for the analysis; 14 of the regressions were for protective factors and 24 for risk factors. To account for the multiple testing, the Holm (1979) method was utilized and adjusts alpha values based on the number of tests conducted. The variables age, grade, gender, race/ethnicity, and parent education were included as covariates in all the logistic and linear regression models (see online supplemental material for covariate reporting). The proportion of missing data for the risk/protective factor variables contained in outcomes ranged from 4.3% to 16.8% (M = 9.6%, SD = 4.3%). Due to the missing data, logistical regressions were conducted using the covariates as predictors and missing or not missing for each risk/protective factor as the dichotomous outcome; results indicated that, in general, students in earlier grades and reporting lower parental education were more likely to have missing data on the outcomes in the analysis. Listwise deletion was used to address the missing data in the larger analysis.

Results

Sociodemographics for Sample/Population and Subgroups

Table 2 contains the unweighted (sample) and weighted (population) estimates for the sociodemographic variables. The mean age of the sample was 14.04 (SD = 2.2) with slightly more females (51.8%) compared with males (47.9%). Most students in the sample identified as White (72.1%) or Latino/a (13.3%), which reflects the state demographics from which the sample was drawn. In addition, 58.7% of youth in the sample reported a religious preference with the Church of Jesus Christ of Latter-day Saints (LDS), whereas an estimated 18.5% reported another religion (e.g., Catholic, Jewish, Protestant), and 22.8% reported having no religious preference. Table 3 contains the weighted sociodemographic variable estimates for the three subgroups of e-cigarette users. Almost 9% of students (i.e., 5.2% infrequent use and 3.4% frequent use) reported using e-cigarettes in the past 30 days. Students in 12th grade reported the highest proportions of infrequent (8.3%) and frequent (7.2%) e-cigarette use across any grade. Similar percentages of males (5.1%) and females (5.3%) were in the infrequent use group, whereas a slightly higher percentage of males was in the frequent use group (3.9% males vs. 3.0% females).

Table 2.

Sociodemographic Variables for Unweighted (Sample) and Weighted (Population) Estimates.

Sociodemographic characteristics	Unweighted estimates		Weighted estimates
Sociodemographic characteristics	n	M (SD), %	N	M (SE), %
Age	48,352	14.04 (2.2)	184,801	14.49 (0.2)
Gender
Male	23,148	47.9%	93,914	50.8%
Female	25,028	51.8%	90,618	49.0%
No response	176	0.4%	268	0.1%
Grade
6th	15,265	31.6%	47,945	25.9%
8th	12,557	30.1%	47,349	25.6%
10th	10,391	21.5%	46,059	24.9%
12th	8,139	16.8%	43,448	23.5%
Race/ethnicity
White	34,624	72.1%	137,717	75.1%
Latino/a	6,392	13.3%	23,934	13.0%
Black/African American	731	1.5%	2,345	1.3%
Asian/Asian American	850	1.8%	3,187	1.7%
Native American	833	1.7%	2,038	1.1%
Native Hawaiian/Pacific Islander	680	1.4%	2,903	1.6%
Multiracial	3,886	8.1%	11,363	6.2%
Parental education
No high school diploma	2,753	6.9%	10,336	6.6%
High school diploma	12,584	31.3%	47,028	30.0%
College degree	24,811	61.8%	99,477	63.4%

Note. Percentages for the sample are valid response percentages within the sociodemographic factor. Weighted (population) estimates were calculated using complex sampling analysis with district as the stratification variable, school as the clustering variable, and sample weights.

Table 3.

Weighted (Population) Estimates of Sociodemographic Characteristics of E-Cigarette Use Subgroups.

Sociodemographic characteristics	Non-e-cigarette users		Infrequent e-cigarette users (1–5 days in the last 30 days)		Frequent e-cigarette users (6+ days in the last 30 days)
Sociodemographic characteristics	N	M (SD), %	N	M (SD), %	N	M (SD), %
Age	168,868	14.35 (0.151)^a	9,610	15.66 (0.115)^b	6,322	16.26 (0.095)^c
Gender
Male	85,476	91.0%^a	4,815	5.1%^a	3,624	3.9%^a
Female	83,146	91.8%^b	4,786	5.3%^a	2,686	3.0%^b
No response	246	91.8%^a,b	9	3.4%^a	13	4.9%^a
Grade
6th	47,161	98.4%^a	657	1.4%^a	128	0.3%^a
8th	44,649	94.3%^b	1,986	4.2%^b	714	1.5%^b
10th	40,353	87.6%^c	3,373	7.3%^c	2,334	5.1%^c
12th	36,705	84.5%^d	3,595	8.3%^c	3,147	7.2%^d
Race/ethnicity
White	126,626	91.9%^a	6,342	4.6%^a	4,749	3.4%^a
Latino/a	21,391	89.4%^b	1,807	7.5%^b	736	3.1%^a
Black/African American	2,124	90.5%^b	151	6.4%^b	71	3.0%^a,b
Asian/Asian American	3,024	94.9%^a	104	3.3%^a	59	1.9%^a
Native American	1,781	87.4%^b	174	8.5%^b	83	4.1%^a,b
Native Hawaiian/Pacific Islander	2,624	90.4%^a,b	189	6.5%^a,b	90	3.1%^a,b
Multiracial	10,024	88.2%^b	821	7.2%^b	519	4.6%^b
Parental education
No high school diploma	9,046	87.5%^a	853	8.3%^a	437	4.2%^a
High school diploma	40,290	85.7%^a	4,044	8.6%^a	2,695	5.7%^a
College degree	93,379	93.7%^b	3,525	3.5%^b	2,574	2.6%^b

Note. Weighted (population) estimates were made using complex sampling analysis with district as the stratification variable, school as the clustering variable, and calculated sample weights. Percentages represent the % of the population estimate of the level of the sociodemographic factor within the e-cigarette use group. ANOVA = analysis of variance.

a,b,c,d

indicates subsets of responses within those categories whose proportion did not significantly differ from each other (p > .05). Chi-square was used for nominal variables, and ANOVA was used for continuous variables.

Differences in Lifetime Substance Use and Age of Initiation

The weighted percentages and odds ratios (ORs) for lifetime substance use (i.e., cigarettes, alcohol, and cannabis) and age of initiation (age < 14 years) across the three e-cigarette subgroups are presented in Table 4. In general, the proportion of students reporting lifetime substance use increased with the level of e-cigarette use in each subgroup. The odds of lifetime cannabis use were 44.72 times higher (p < .001) for students frequently using e-cigarettes compared with those not using e-cigarettes. The odds of lifetime alcohol (OR = 27.35, p < .001) and cigarette (OR = 20.69, p < .001) use for those in the frequent e-cigarette group were more than 20 times higher compared with those not using e-cigarettes. In addition to lifetime use of substances, larger proportions of students who used e-cigarettes also reported using other substances, prior to age 14 years, compared with those not using e-cigarettes. For example, the odds of using cannabis before age 14 years were 17.15 times higher (p < .001) for students who reported frequent e-cigarette use compared with those who did not report e-cigarette use. The odds of alcohol (OR = 7.08, p < .001) and cigarette (OR = 11.71, p < .001) use were at least more than seven times higher for frequent e-cigarette users compared with students not using e-cigarettes. Taken together, these findings suggest that students reporting the use of e-cigarettes were more likely to have used another substance in their lifetime (i.e., alcohol, cannabis, tobacco) and used the substance(s) prior to age 14 years compared with those students who did not report using e-cigarettes.

Table 4.

Weighted (Population) Comparison of Lifetime, Early Initiation, and Risk and Protective Factors Between E-Cigarette Use Study Subgroups.

Factor	Non-e-cigarette users (N = 168,868)	Infrequent e-cigarette users (1–5 days in the last 30 days; N = 9,610)	Frequent e-cigarette users (6+ days in the last 30 days; N = 6,322)	Infrequent users vs. nonusers	Frequent users vs. nonusers	Frequent users vs. infrequent users	n
Factor	M, % (SE)	M, % (SE)	M, % (SE)	Effect size [95% CI]	Effect size [95% CI]	Effect size [95% CI]	n
Lifetime substance use
Cigarette^a,b	5.0% (0.3%)	42.1% (1.3%)	60.2% (1.8%)	10.710*** [8.726, 11.854]	20.692*** [17.427, 24.568]	2.035*** [1.725, 2.400]	48,052
Alcohol^a,b	15.8% (0.7%)	74.8% (1.3%)	87.0% (1.3%)	12.917*** [11.190, 14.911]	27.354*** [21.497, 34.807]	2.118*** [1.616, 2.775]	48,022
Cannabis^a,b	7.1% (0.5%)	60.5% (1.5%)	81.3% (1.9%)	16.139*** [13.571, 19.192]	44.720*** [32.270, 61.972]	2.771*** [2.107, 3.644]	48,078
Early initiation (age < 14 years)
Cigarette^a,b	2.9% (0.1%)	22.4% (1.2%)	28.3% (1.8%)	7.573*** [6.353, 9.028]	11.705*** [9.464, 14.476]	1.546*** [1.258, 1.898]	48,052
Alcohol^a,b	8.2% (0.3%)	32.3% (1.8%)	35.5% (2.2%)	4.980*** [4.258, 5.824]	7.077*** [5.680, 8.817]	1.421*** [1.171, 1.725]	48,022
Cannabis^a,b	2.4% (0.2%)	22.9% (1.4%)	31.4% (2.4%)	10.028*** [8.402, 11.968]	17.153*** [13.227, 22.243]	1.711*** [1.353, 2.163]	48,078
Individual
Perceived risk of harm from e-cigarette use^c,d	3.148 (0.012)	2.195 (0.035)	1.983 (0.029)	−0.913*** [−0.956, −0.869]	−1.115*** [−1.103, −1.084]	−0.202*** [−0.214, −0.190]	47,509
Perceived risk of harm from alcohol^c,d	3.365 (0.009)	2.979 (0.025)	2.695 (0.023)	−0.410*** [−0.441, −0.376]	−0.712*** [−0.743, −0.681]	−0.302*** [−0.305, −0.299]
Perceived risk of harm from cannabis^c,d	3.159 (0.013)	2.298 (0.032)	2.047 (0.029)	−0.853*** [−0.889, −0.816]	−1.100*** [−1.130, −1.070]	−0.248*** [−0.254, −0.241]	46,764
Rebelliousness^b,c	1.509 (0.007)	1.799 (0.025)	1.817 (0.026)	0.543*** [0.477, 0.608]	0.575*** [0.506, 0.645]	0.033 [0.029, 0.036]	22,599^e
Favorable attitudes toward antisocial behavior^b,c	1.432 (0.005)	1.716 (0.015)	1.749 (0.021)	0.616*** [0.572, 0.659]	0.685*** [0.618, 0.753]	0.070 [0.045, 0.094]	47,367
Favorable attitudes toward substance use^b,c	1.220 (0.006)	1.845 (0.021)	2.181 (0.026)	1.235*** [1.180, 1.290]	1.899*** [1.824, 1.975]	0.664*** [0.643, 0.686]	47,659
Belief in a moral order^c,d	3.283 (0.008)	2.879 (0.028)	2.766 (0.032)	−0.657*** [−0.723, −0.591]	−0.840*** [−0.917, −0.763]	−0.183 [−0.194, −0.172]	21,997^e
Prosocial involvement^c,d	1.946 (0.009)	1.754 (0.023)	1.656 (0.029)	−0.246*** [−0.280, −0.211]	−0.408*** [−0.461, −0.355]	−0.162*** [−0.181, −0.144]	47,242
Peer
Peer substance use^b,c	1.238 (0.009)	2.146 (0.037)	2.680 (0.071)	1.389*** [1.306, 1.471]	2.205*** [2.019, 2.391]	0.817*** [0.713, 0.920]	23,186^e
Peer favorable attitudes toward drug use^b,c	1.243 (0.007)	1.898 (0.024)	2.202 (0.032)	1.141*** [1.084, 1.198]	1.672*** [1.586, 1.757]	0.531*** [0.502, 0.559]	47,495
Being perceived as “cool” for cigarette use^b,f	1.380 (0.008)	1.597 (0.041)	1.742 (0.046)	0.258*** [0.162, 0.334]	0.429*** [0.342, 0.517]	0.172 [0.161, 0.183]	22,210^e
Being perceived as “cool” for alcohol use^b,f	1.481 (0.013)	1.937 (0.041)	2.242 (0.066)	0.484*** [0.425, 0.542]	0.806*** [0.697, 0.916]	0.323*** [0.271, 0.374]	22,150^e
Being perceived as “cool” for cannabis use^b,f	1.577 (0.017)	2.342 (0.054)	2.518 (0.067)	0.707*** [0.638, 0.775]	0.869*** [0.778, 0.960]	0.163 [0.140, 0.185]	22,100^e
Being perceived as “cool” for substance use^b,f	1.480 (0.012)	1.959 (0.035)	2.166 (0.049)	0.547*** [0.494, 0.600]	0.784*** [0.700, 0.868]	0.237*** [0.206, 0.267]	21,969^e
Interactions with antisocial peers^b,f	1.101 (0.003)	1.311 (0.022)	1.489 (0.034)	0.643*** [0.529, 0.757]	1.191*** [1.005, 1.377]	0.548*** [0.476, 0.620]	23,147^e
Being perceived as “cool” for prosocial involvement^d,f	3.338 (0.013)	3.063 (0.059)	3.059 (0.057)	−0.241*** [−0.320, −0.162]	−0.244*** [−0.319, −0.170]	−0.004 [−0.008, 0.001]	21,894^e
Interactions with prosocial peers^d,f	3.593 (0.016)	2.999 (0.039)	2.784 (0.049)	−0.552*** [−0.593, −0.511]	−0.752*** [−0.811, −0.693]	−0.200*** [−0.218, −0.181]	22,835^e
Family
Family history of alcohol and/or drug use^a,b	39.8% (0.9%)	81.2% (1.7%)	84.6% (2.3%)	4.851*** [3.719, 6.327]	6.188*** [4.011, 9.545]	1.276 [0.769, 2.116]	22,753^e
Family conflict^b,c	2.063 (0.008)	2.377 (0.023)	2.373 (0.035)	0.421*** [0.381, 0.461]	0.415*** [0.343, 0.488]	−0.006 [−0.038, 0.027]	47,289
Family management^c,d	3.423 (0.007)	3.053 (0.028)	2.964 (0.025)	−0.609*** [−0.677, −0.541]	−0.756*** [−0.816, −0.687]	−0.147 [−0.139, −0.155]	21,589^e
Parental favorable attitudes toward substance use^b,c	1.080 (0.003)	1.336 (0.015)	1.387 (0.019)	0.795*** [0.718, 0.873]	0.954*** [0.851, 1.057]	0.158 [0.133, 0.183]	47,471
Family attachment^c,d	3.236 (0.009)	2.883 (0.036)	2.835 (0.045)	−0.478*** [−0.549, 0.407]	−0.543*** [−0.640, −0.446]	−0.065 [−0.091, −0.039]	22,828^e
Prosocial opportunities^c,d	3.227 (0.009)	2.890 (0.037)	2.808 (0.050)	−0.456*** [−0.530, −0.382]	−0.566*** [−0.673, −0.458]	−0.110 [−0.144, −0.076]	22,927^e
Prosocial rewards^c,d	3.215 (0.008)	2.918 (0.032)	2.881 (0.041)	−0.443*** [−0.512, −0.374]	−0.498*** [−0.594, −0.403]	−0.055 [−0.082, −0.029]	22,003^e
School
Academic failure^b,c	1.826 (0.009)	2.197 (0.023)	2.316 (0.026)	0.506*** [0.468, 0.544]	0.668*** [0.623, 0.713]	0.161*** [0.154, 0.169]	46,518
Low commitment to school^b,f	2.352 (0.007)	2.763 (0.021)	2.911 (0.023)	0.600*** [0.560, 0.641]	0.816*** [0.768, 0.863]	0.216*** [0.209, 0.222]	46,630
Prosocial opportunities^c,d	3.020 (0.006)	2.900 (0.015)	2.868 (0.014)	−0.264*** [−0.304, −0.224]	−0.335*** [−0.369, −0.302]	−0.071 [−0.078, −0.065]	46,914
Prosocial rewards^c,d	2.813 (0.009)	2.592 (0.022)	2.595 (0.020)	−0.370*** [−0.413, −0.327]	−0.364*** [−0.401, −0.328]	0.005 [−0.001, 0.012]	46,637
Community
Laws and norms favorable to substance use^b,c	1.737 (0.009)	2.077 (0.026)	2.087 (0.029)	0.613*** [0.550, 0.676]	0.631*** [0.559, 0.702]	0.017 [0.009, 0.026]	22,100^e
Perceived availability of substances^b,c	1.797 (0.013)	2.711 (0.032)	2.927 (0.049)	0.990*** [0.948, 1.032]	1.224*** [1.146, 1.301]	0.234*** [0.198, 0.269]	22,509^e
Low neighborhood attachment^b,c	1.841 (0.010)	2.182 (0.030)	2.183 (0.049)	0.429*** [0.380, 0.477]	0.430*** [0.332, 0.527]	0.001 [−0.048, 0.050]	22,743^e
Rewards for prosocial involvement^c,d	2.564 (0.016)	2.117 (0.030)	2.004 (0.040)	−0.490*** [−0.521, −0.460]	−0.614*** [−0.666, 0.563]	−0.124 [−0.145, −0.103]	23,056^e

Note. Percentages are valid response percentages within e-cigarette use subgroups after correction for complex sampling. Means are estimated marginal means adjusted for sociodemographic characteristics (age, grade, gender, race/ethnicity, and caregiver education) and complex sampling. Effect sizes are odds ratio for dichotomous items and estimated mean differences in SD units for continuous variables after correction for complex sampling; n is unweighted valid item responses. CI = confidence interval.

Dichotomous item. ^b Risk factor. ^c Maximum score of 4 on scale. ^d Protective factor. ^e Sample “n” is smaller because item was included on only one version of the survey (i.e., A or B). ^f Maximum score of 5 on scale.

***

p < .001, the alpha level adjusted for multiple testing using the Holm (1979) method.

Differences in Levels of Risk and Protective Factors

The e-cigarette use subgroups generally predicted differing levels of risk and protective factors in the five domains (see Table 4). Students in the frequent and infrequent use subgroups reported lower levels of protection and higher levels of risk compared with students in the nonuse subgroup. The greatest difference in effect size (ES) was observed between students in the frequent use subgroup compared with those in the nonuse subgroup. One of the largest ESs (ES = 2.21, p < .001) was found for students who frequently used e-cigarettes who also indicated that their peers used substances. In addition, students who frequently used e-cigarettes were more likely to report higher individual (ES = 1.90, p < .001), peer (ES = 1.67, p < .001), and parental (ES = 0.95, p < .001) favorable attitudes toward substance use and lower perceived risk of harm from e-cigarette use (ES = −1.12, p < .001) compared with those who did not use e-cigarettes. Of note, significant differences were observed in less than half (44%) of the risk and protective factors for adolescents in the infrequent versus frequent e-cigarette subgroups; in particular, the family domain did not contain any significant differences for risk and protective factors between the infrequent and frequent use groups. Compared with the individual or peer domains, the school and community domains contained the least number of significant differences for risk and protective factors between the infrequent and frequent use groups.

Discussion

We conducted a comprehensive examination of e-cigarette use in relation to risk and protective factors for a large school-based sample of adolescents. The findings from our three research questions indicated that the frequency of adolescent e-cigarette use predicted differences in lifetime substance use, early initiation of substance use, and the levels of risk and protective factors. Findings from the present study add to the literature by providing a more comprehensive examination of adolescent e-cigarette use and associated risk and protective factors than what has been published to date. In addition, the findings have important implications for preventing e-cigarette use in adolescents.

The findings from our study indicate that the frequency of e-cigarette use in adolescents has a predictive value in understanding the lifetime and early initiation of using other substances. For example, frequent and infrequent users of e-cigarettes had higher odds of cannabis, alcohol, and cigarette use in their lifetime compared with adolescents who did not use e-cigarettes. Frequent e-cigarette users had the highest odds of lifetime cannabis use (i.e., almost 45 times higher) compared with nonusers of e-cigarettes. The findings also indicate that frequent and infrequent e-cigarette users had higher odds of using other substances before the age of 14 years compared with adolescents who did not use e-cigarettes. In fact, the odds of cannabis use before the age of 14 years were more than 17 times higher for frequent e-cigarette users compared with nonusers. Taken together, these findings imply that adolescents who use e-cigarettes have a high likelihood of cannabis use as well as other substances; thus, e-cigarette prevention efforts should consider targeting multiple substances, especially cannabis, due to the use of this substance with e-cigarette devices (Dai & Siahpush, 2019; Morean et al., 2015).

The frequency of adolescent e-cigarette use also predicted differences in the levels of risk and protective factors across the five socializing domains. Frequent and infrequent e-cigarette users consistently displayed higher levels of risk factors (e.g., favorable attitude toward substance use) and lower levels of protective factors (e.g., perceived risk of harm from e-cigarette use) compared with nonusers. The largest differences in ES for risk factors were found between frequent and nonusers of e-cigarettes for family history of substance use, peer substance use, favorable attitudes toward substance use (i.e., individual, peer, and family domains), perceived availability of substances, interactions with antisocial peers, and low commitment to school. These findings are largely consistent with the literature on risk factors for alcohol, cannabis, and tobacco (see Flay et al., 1999; Hawkins et al., 1992; Meier et al., 2019). Perhaps not surprisingly, the risk factor with the largest ES distinguishing frequent users and nonusers of e-cigarettes was having peers who used substances (i.e., peer domain); the influence of peer use is a consistent finding for other substances (see Brook et al., 1999; Chassin et al., 2002; van den Bree et al., 2004). Large ESs were also found between frequent users and nonusers of e-cigarettes for risk factors in the individual, peer, and parental domains. Specifically, students who frequently used e-cigarettes reported that their attitudes (i.e., individual domain) and others close to them (i.e., peer and parent domains) were more favorable toward the use of substances compared with students who did not use e-cigarettes. These findings are consistent with prior research on adolescent substance use (see Guo et al., 2001; Mason et al., 2014; McDermott, 1984) indicating that individual attitudes and the perspectives of those around them (i.e., peers, parents) may influence use behavior. Considering these findings, prevention efforts should continue targeting the favorable attitudes adolescents may have for using substances including e-cigarettes within the context of preventive interventions.

The protective factors with the largest differences in ES were found between frequent users and nonusers of e-cigarettes for perceived risk of harm (i.e., individual domain for e-cigarettes, cannabis, and alcohol), belief in a moral order, interactions with prosocial peers, and rewards for prosocial involvement. We highlight that students who did not use e-cigarettes in the past 30 days reported higher levels of perceived risk (i.e., physically or other ways) for the use of e-cigarettes, cannabis, or alcohol compared with youth who frequently used e-cigarettes. These findings are generally consistent with prior research indicating that youth who use e-cigarettes tend to view them as less harmful compared with other substances including combustible cigarettes (Roditis et al., 2016; Tsai et al., 2018). The frequent users of e-cigarettes also indicated significantly less belief in a moral order (i.e., individual domain: belief in system of right and wrong) compared with nonusers of e-cigarettes; this finding is consistent with adolescent alcohol use (Monahan et al., 2014). Finally, the adolescents in the frequent e-cigarette use group reported significantly less interactions with prosocial peers compared with those in the nonuse group; this is consistent with recent findings for mixed substance use that includes alcohol, cannabis, or tobacco (Walters, 2020). Overall, perceived risk of harm in the individual domain appears to be one of the most salient protective factors for adolescent e-cigarette use and should be integrated into preventive interventions similar to those of other substances such as cannabis (Terry-McElrath et al., 2017).

Findings from the current study provide important implications for the prevention of e-cigarette use in adolescents. First, our findings extend the risk and protective factor literature found for other substances (i.e., alcohol, cannabis, tobacco) to the use of e-cigarettes for adolescents. In other words, there is likely shared risk and protective factors across different substances and prevention efforts may have broader impact by targeting multiple substances, including e-cigarettes, instead of just one. Second, one of the most salient risk factors identified in the current study is attitudes in favor of using substances (i.e., individual, peer, familial domains). Thus, prevention efforts that allow adolescents to examine their favorable attitudes toward e-cigarettes and other substances as well as exploring alternative attitudes that support nonuse may prove useful in buffering this salient risk factor. Third, the perceived risk of harm from using e-cigarettes as well as cannabis and alcohol was a salient protective factor. We suggest that preventive interventions integrate accurate information about the known risks of e-cigarette use, recognizing that this literature is still evolving. It is well known that effective prevention efforts should not only reduce risk factors but also promote protective factors (Catalano & Hawkins, 1996; Hawkins et al., 1992). Therefore, we suggest that the risk and protective factors from the last two points (i.e., attitudes and perceived risk) should be considered together in the context of preventing adolescent e-cigarette use. For example, providing youth with accurate information about the potential harms of e-cigarettes could influence their perceptions of risk as well as their attitudes toward use. It is important to note, however, that we do not know the long-term effects of e-cigarette use compared with other substances such as alcohol, cannabis, and tobacco; therefore, being honest with adolescents about what is currently known or not known regarding e-cigarettes as well as other substances is important for successful prevention efforts. In fact, some research indicates that adolescents express wanting accurate information on the physical effects of using substances (see Morton et al., 2015). We suggest that directly providing youth with accurate information on the effects of substances could be used in concert with environmental strategies (e.g., banning flavors, raising age limits) to provide comprehensive prevention approaches across multiple domains.

As with all studies, the present study has certain limitations that need to be considered. First, data were collected in the state of Utah and may not generalize to students in other states or across the nation. For example, the e-cigarette use rates in particular, and substance use rates generally, are higher in national studies (Cullen et al., 2019; Gentzke et al., 2019; Miech et al., 2019) compared with those presented in this article. However, one of the concerns across surveys of e-cigarettes is a lack of a consistency in measurement (see Weaver et al., 2018) and this may lead to inaccurate comparisons of use rates across studies. Thus, the lower use rates in the present study may be, in part, due to the difficulty in trying to make comparisons across survey studies that use different measurement approaches. However, the lower overall use rates could be because almost 59% of the adolescents in the study reported an LDS religious preference, which is the predominant religion in the state of Utah. Prior research suggests that religiosity may serve as a protective factor for adolescent substance use (Hodge et al., 2001; Mason et al., 2012) and may help to explain lower overall use rates in the present study. A second potential limitation of the present study is that data were collected from students through a self-report survey method and not corroborated by other sources such as peers or parents. Nevertheless, the data were collected in ways to mitigate underreporting, overreporting, or false reporting by allowing respondents to retain their anonymity and excluding data that failed validity checks. Similarly, national surveys, such as Monitoring the Future (2020), utilize self-report methods as those used in the current study. A third potential limitation is that five of the risk and protective factor subscales had alpha reliabilities of less than .70, and future researchers should consider ways to increase the reliability for these subscales. Finally, data in the present study are cross-sectional and do not allow for examination of temporal relations among risk and protective factors. We recommend that longitudinal data are needed to assess the trajectories of adolescent risk and protective factors for e-cigarette use across time.

To date, a comprehensive examination of risk and protective factors for adolescent e-cigarette use has been lacking in the literature. Prior studies on e-cigarettes have been limited in scope and conducted with modest sample sizes (see Dunbar et al., 2017; Kwon et al., 2018; Wills et al., 2015). Our study is one of the first to examine a comprehensive set of risk and protective factors across five domains with a large adolescent school-based sample. The findings in the present study are also consistent with the SDM (Catalano & Hawkins, 1996) in that risk and protective factors can apply to multiple substances including e-cigarettes. The findings from this study provide researchers, practitioners, and policy makers with information on how the risk and protective factors identified for other substances (i.e., alcohol, cannabis, and tobacco) extend to adolescent e-cigarette use.

Supplemental Material

sj-pdf-1-prv-10.1177_2632077020980734 – Supplemental material for Adolescent Risk and Protective Factors for the Use of Electronic Cigarettes

Supplemental material, sj-pdf-1-prv-10.1177_2632077020980734 for Adolescent Risk and Protective Factors for the Use of Electronic Cigarettes by Jason J. Burrow-Sánchez and Benjamin R. Ratcliff in Journal of Prevention and Health Promotion

Footnotes

Acknowledgements

We acknowledge and thank colleagues at Bach Harrison, LLC, the Utah Division of Substance Abuse and Mental Health, and the Utah Department of Health for their assistance with this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Jason J. Burrow-Sánchez is a Professor of Counseling Psychology in the Department of Educational Psychology at the University of Utah. His reseach interests include the prevention and treatment of substance use in adolescents.

Benjamin R. Ratcliff is a doctoral candidate within the Counseling Psychology program at the University of Utah. His research interests include the risk and protective factors of adolescent substance use, the impact of childhood trauma, and prevention.

References

Arthur

M. W.

Briney

J. S.

Hawkins

J. D.

Abbott

R. D.

Brooke-Weiss

B. L.

Catalano

R. F.

(2007). Measuring risk and protection in communities using the Communities That Care Youth Survey. Evaluation and Program Planning, 30(2), 197–211. https://doi.org/10.1016/j.evalprogplan.2007.01.009

Arthur

M. W.

Hawkins

J. D.

Pollard

J. A.

Catalano

R. F.

Baglioni

A. J.

Jr. (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors. The Communities That Care Youth Survey. Evaluation Review, 26(6), 575–601. https://doi.org/10.1177/0193841x0202600601

Bach Harrison. (2017). 2017 prevention needs assessment survey results. https://drive.google.com/file/d/1AdArgzCAe1HBtZK5vSGATRRjWPRdXmdU/view

Bhatta

D. N.

Glantz

S. A.

(2019). Association of e-cigarette use with respiratory disease among adults: A longitudinal analysis. American Journal of Preventive Medicine, 58(2), 182–190. https://doi.org/10.1016/j.amepre.2019.07.028

Blount

B. C.

Karwowski

M. P.

Shields

P. G.

Morel-Espinosa

Valentin-Blasini

Gardner

… Pirkle

J. L.

(2019). Vitamin E acetate in bronchoalveolar-lavage fluid associated with EVALI. The New England Journal of Medicine, 382, 697–705. https://doi.org/10.1056/NEJMoa1916433

Briney

J. S.

Brown

E. C.

Hawkins

J. D.

Arthur

M. W.

(2012). Predictive validity of established cut points for risk and protective factor scales from the Communities That Care Youth Survey. The Journal of Primary Prevention, 33(5–6), 249–258.

Brook

J. S.

Kessler

R. C.

Cohen

(1999). The onset of marijuana use from preadolescence and early adolescence to young adulthood. Development and Psychopathology, 11(4), 901–914. https://doi.org/10.1017/s0954579499002370

Catalano

R. F.

Hawkins

J. D.

(1996). The social development model: A theory of antisocial behavior. In Hawkins

J. D.

(Ed.), Delinquency and crime (pp. 149–197). Cambridge University Press. https://doi.org/10.1007%2Fs10935-012-0280-1

Chaffee

B. W.

Watkins

S. L.

Glantz

S. A.

(2018). Electronic cigarette use and progression from experimentation to established smoking. Pediatrics, 141(4), Article e20173594. https://doi.org/10.1542/peds.2017-3594

10.

Chassin

Pitts

S. C.

Prost

(2002). Binge drinking trajectories from adolescence to emerging adulthood in a high-risk sample: Predictors and substance abuse outcomes. Journal of Consulting and Clinical Psychology, 70(1), 67–78. https://psycnet.apa.org/doi/10.1037/0022-006X.70.1.67

11.

Cullen

K. A.

Gentzke

A. S.

Sawdey

M. D.

Chang

J. T.

Anic

G. M.

Wang

T. W.

… King

B. A.

(2019). E-cigarette use among youth in the United States, 2019. Journal of the American Medical Association, 322(21), 2095–2103. https://doi.org/10.1001/jama.2019.18387

12.

Dai

Siahpush

. (2019). Use of e-cigarettes for nicotine, marijuana, and just flavoring among U.S. youth. American Journal of Preventive Medicine, 58(2), 244–249. https://doi.org/10.1016/j.amepre.2019.09.006

13.

Dunbar

M. S.

Tucker

J. S.

Ewing

B. A.

Pedersen

E. R.

Miles

J. N. V.

Shih

R. A.

D’Amico

E. J.

(2017). Frequency of e-cigarette use, health status, and risk and protective health behaviors in adolescents. Journal of Addiction Medicine, 11(1), 55–62. https://doi.org/10.1097/ADM.0000000000000272

14.

Feinberg

M. E.

Ridenour

T. A.

Greenberg

M. T.

(2007). Aggregating indices of risk and protection for adolescent behavior problems: The Communities That Care Youth Survey. Journal of Adolescent Health, 40(6), 506–513. https://doi.org/10.1016/j.jadohealth.2006.09.002

15.

Flay

B. R.

Petraitis

F. B.

(1999). Psychosocial risk and protective factors for adolescent tobacco use. Nicotine & Tobacco Research, 1, S59–S65. https://doi.org/10.1080/14622299050011611

16.

Gentzke

A. S.

Creamer

Cullen

K. A.

Ambrose

B. K.

Willis

Jamal

King

B. A.

(2019). Vital signs: Tobacco use among middle and high school students—United States, 2011-2018. Morbidity and Mortality Weekly Report, 68(6), 157–164. https://www.cdc.gov/mmwr/volumes/68/wr/mm6806e1.htm?s_cid=mm6806e1_w

17.

Glaser

R. R.

Horn

M. L. V.

Arthur

M. W.

Hawkins

J. D.

Catalano

R. F.

(2005). Measurement properties of the Communities That Care® Youth Survey across demographic groups. Journal of Quantitative Criminology, 21(1), 73–102. https://psycnet.apa.org/doi/10.1007/s10940-004-1788-1

18.

Goniewicz

M. L.

Gupta

Lee

Y. H.

Reinhardt

Kim

… Sobczak

(2015). Nicotine levels in electronic cigarette refill solutions: A comparative analysis of products from the U.S., Korea, and Poland. International Journal of Drug Policy, 26(6), 583–588. https://doi.org/10.1016/j.drugpo.2015.01.020

19.

Guo

Hawkins

J. D.

Hill

K. G.

Abbott

R. D.

(2001). Childhood and adolescent predictors of alcohol abuse and dependence in young adulthood. Journal of Studies on Alcohol, 62(6), 754–762. https://doi.org/10.15288/jsa.2001.62.754

20.

Hawkins

J. D.

Catalano

Miller

J. Y.

(1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105. http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=9210193947&site=ehost-live

21.

Hawkins

J. D.

Lishner

D. M.

Catalano

R. F.

Howard

M. O.

(1986). Childhood predictors of adolescent substance abuse. Journal of Children in Contemporary Society, 18(1–2), 11–48. https://doi.org/10.1300/J274v18n01_03

22.

Hawkins

J. D.

Weis

J. G.

(1985). The social development model: An integrated approach to delinquency prevention. Journal of Primary Prevention, 6(2), 73–97. https://doi.org/10.1007/BF01325432

23.

Hodge

D. R.

Cardenas

Montoya

(2001). Substance use: Spirituality and religious participation as protective factors among rural youths. Social Work Research, 25(3), 153–161. https://doi.org/10.1093/swr/25.3.153

24.

Holm

(1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70. http://www.jstor.org/stable/4615733

25.

Jackson

K. M.

Sher

K. J.

Schulenberg

J. E.

(2005). Conjoint developmental trajectories of young adult alcohol and tobacco use. Journal of Abnormal Psychology, 114(4), 612–626. https://doi.org/10.1037/0021-843x.114.4.612

26.

Kim

B. K. E.

Oesterle

Catalano

R. F.

Hawkins

J. D.

(2015). Change in protective factors across adolescent development. Journal of Applied Developmental Psychology, 40, 26–37. https://doi.org/10.1016/j.appdev.2015.04.006

27.

Kuttler

Schwendemann

Bitzer

E. M.

(2015). Familial risk and protective factors in alcohol intoxicated adolescents: Psychometric evaluation of the family domain of the Communities That Care Youth Survey (CTC) and a new short version of the Childhood Trauma Questionnaire (CTQ). BMC Pediatrics, 15(1), Article 191. https://doi.org/10.1186/s12887-015-0471-z

28.

Kwon

Seo

D.-C.

Lin

H.-C.

Chen

(2018). Predictors of youth e-cigarette use susceptibility in a U.S. nationally representative sample. Addictive Behaviors, 82, 79–85. https://doi.org/10.1016/j.addbeh.2018.02.026

29.

Mason

M. J.

Mennis

Linker

Bares

Zaharakis

(2014). Peer attitudes effects on adolescent substance use: The moderating role of race and gender. Prevention Science, 15(1), 56–64. https://doi.org/10.1007/s11121-012-0353-7

30.

Mason

M. J.

Schmidt

Mennis

(2012). Dimensions of religiosity and access to religious social capital: Correlates with substance use among urban adolescents. The Journal of Primary Prevention, 33(5–6), 229–237. https://doi.org/10.1007/s10935-012-0283-y

31.

McDermott

(1984). The relationship of parental drug use and parents’ attitude concerning adolescent drug use to adolescent drug use. Adolescence, 19(73), 89–97.

32.

Meier

M. H.

Docherty

Leischow

S. J.

Grimm

K. J.

Pardini

(2019). Cannabis concentrate use in adolescents. Pediatrics, 144(3), Article e20190338. https://doi.org/10.1542/peds.2019-0338

33.

Miech

Johnston

O’Malley

P. M.

Bachman

J. G.

Patrick

M. E.

(2019). Trends in adolescent vaping, 2017–2019. The New England Journal of Medicine, 381(15), 1490–1491. https://doi.org/10.1056/NEJMc1910739

34.

Monahan

K. C.

Oesterle

Rhew

Hawkins

J. D.

(2014). The relation between risk and protective factors for problem behaviors and depressive symptoms, antisocial behavior, and alcohol use in adolescence. Journal of Community Psychology, 42(5), 621–638. https://doi.org/10.1002/jcop.21642

35.

Monitoring the Future. (2020). Monitoring the future: A continuing study of American youth. http://www.monitoringthefuture.org/

36.

Morean

M. E.

Kong

Camenga

D. R.

Cavallo

D. A.

Krishnan-Sarin

(2015). High school students’ use of electronic cigarettes to vaporize cannabis. Pediatrics, 136(4), 611–616. https://doi.org/10.1542/peds.2015-1727

37.

Morton

C. M.

Hoefinger

Linn-Walton

Aikins

Falkin

G. P.

(2015). What are youth asking about drugs? A report of NIDA drug facts chat day. Journal of Drug Education, 45(3–4), 195–210. https://doi.org/10.1177/0047237915622084

38.

Omaiye

E. E.

Cordova

Davis

Talbot

(2017). Counterfeit electronic cigarette products with mislabeled nicotine concentrations. Tobacco Regulatory Science, 3(3), 347–357. https://doi.org/10.18001/trs.3.3.10

39.

Raymond

B. H.

Collette-Merrill

Harrison

R. G.

Jarvis

Rasmussen

R. J.

(2018). The nicotine content of a sample of e-cigarette liquid manufactured in the United States. Journal of Addiction Medicine, 12(2), 127–131. https://doi.org/10.1097/adm.0000000000000376

40.

Roditis

Delucchi

Cash

Halpern-Felsher

(2016). Adolescents’ perceptions of health risks, social risks, and benefits differ across tobacco products. Journal of Adolescent Health, 58(5), 558–566. https://doi.org/10.1016/j.jadohealth.2016.01.012

41.

Scal

Ireland

Borowsky

I. W.

(2003). Smoking among American adolescents: A risk and protective factor analysis. Journal of Community Health, 28(2), 79–97. https://doi.org/10.1023/A:1022691212793

42.

Schier

J. G.

Meiman

J. G.

Layden

Mikosz

C. A.

VanFrank

King

B. A.

…Meaney-Delman

(2019). Severe pulmonary disease associated with electronic-cigarette-product use-interim guidance. Morbidity and Mortality Weekly Report, 68(36), 787–790. https://www.cdc.gov/mmwr/volumes/68/wr/mm6836e2.htm

43.

Statistical Package for the Social Sciences. (2019). IBM SPSS statistics (Version 26). International Business Machines.

44.

Terry-McElrath

Y. M.

O’Malley

P. M.

Patrick

M. E.

Miech

R. A.

(2017). Risk is still relevant: Time-varying associations between perceived risk and marijuana use among US 12th grade students from 1991 to 2016. Addictive Behavior, 74, 13–19. https://doi.org/10.1016/j.addbeh.2017.05.026

45.

Tsai

Walton

Coleman

B. N.

Sharapova

S. R.

Johnson

S. E.

Kennedy

S. M.

Caraballo

R. S.

(2018). Reasons for electronic cigarette use among middle and high school students—National Youth Tobacco Survey, United States, 2016. https://www.cdc.gov/mmwr/volumes/67/wr/mm6706a5.htm

46.

Utah Department of Human Services. (2019). SHARP survey. Utah Department of Human Services, Substance Abuse and Mental Health. https://dsamh.utah.gov/reports/sharp-survey

47.

van den Bree

M. B.

Whitmer

M. D.

Pickworth

W. B

. (2004). Predictors of smoking development in a population-based sample of adolescents: A prospective study. Journal of Adolescent Health, 35(3), 172–181. https://doi.org/10.1016/j.jadohealth.2003.09.021

48.

Walters

G. D.

(2020). Prosocial peers as risk, protective, and promotive factors for the prevention of delinquency and drug use. Journal of Youth and Adolescence, 49(3), 618–630. https://doi.org/10.1007/s10964-019-01058-3

49.

Weaver

S. R.

Kim

Glasser

A. M.

Sutfin

E. L.

Barrington-Trimis

Payne

T. J.

… Loukas

(2018). Establishing consensus on survey measures for electronic nicotine and non-nicotine delivery system use: Current challenges and considerations for researchers. Addictive Behaviors, 79, 203–212. https://doi.org/10.1016/j.addbeh.2017.11.016

50.

Wills

T. A.

Knight

Williams

R. J.

Pagano

Sargent

J. D.

(2015). Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics, 135(1), e43–e51. https://doi.org/10.1542/peds.2014-0760

51.

Yuan

Cross

S. J.

Loughlin

S. E.

Leslie

F. M.

(2015). Nicotine and the adolescent brain. The Journal of Physiology, 593(16), 3397–3412. https://doi.org/10.1113/JP270492

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB