RESEARCH ARTICLE
Victims of Bullying and Tobacco Use Behaviors in Adolescents: Differences Between Bullied at School, Electronically, or Both KATHLEEN R. CASE, MPHa MARIA COOPER, PhDb MELISA CREAMER, PhD, MPHc DALE MANTEY, MPAd STEVEN KELDER, PhD, MPHe
ABSTRACT BACKGROUND: Being a victim of bullying is associated with greater risk of youth substance use; however, research specifically examining whether tobacco use behaviors differ among adolescents who were bullied at school only, electronically only, or both at school and electronically is limited. METHODS: We examined the associations between being a victim of bullying (at school only, electronically only, or both at school and electronically) and use of tobacco products using data from the 2013 national Youth Risk Behavior Survey. RESULTS: Girls who were bullied both at school and electronically reported the highest odds of all tobacco use behaviors (ever use of cigarettes, current use of cigarettes, and current use of any tobacco product) as compared with girls who were not bullied after adjusting for covariates. Conversely, for boys, only the association between being bullied electronically only and ever use of cigarettes remained significant after adjusting for covariates. CONCLUSIONS: Results from this study indicate that electronic bullying may differentially influence the odds of tobacco use in high school students as compared with bullying that occurs at school only. Confirmation of these findings could inform interventions to reduce both bullying and tobacco use in high school. Keywords: tobacco; smoking; bullying; Youth Risk Behavior Survey. Citation: Case KR, Cooper M, Creamer M, Mantey D, Kelder S. Victims of bullying and tobacco use behaviors in adolescents: differences between bullied at school, electronically, or both. J Sch Health. 2016; 86: 832-840. Received on August 2, 2015 Accepted on May 16, 2016
the age of 18.3 According to 2013 data, 41% of high school students reported ever smoking cigarettes and 16% reported current smoking.4 Additionally in 2014, 24.6% of high school students reported current use of any tobacco product, which included cigarettes, cigars or cigarillos, and smokeless tobacco.5 While significant gains have been made in reducing the prevalence of cigarette smoking among youth, nearly half of 12th
A
ccording to the Centers for Disease Control and Prevention (CDC), tobacco use is the number one cause of preventable deaths in the United States.1,2 As noted in the literature, tobacco use is implicated in a wide range of health issues from many forms of cancer to type 2 diabetes.2 Tobacco use habits are typically established in adolescence, with 88% of adults who reported ever smoking daily initiating smoking before
a Postdoctoral Trainee, (
[email protected]), University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701. bPostdoctoral Fellow, (
[email protected], University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701. c Faculty Associate, (
[email protected]), Department of Health Promotion and Behavioral Sciences, University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701. d Predoctoral Trainee, (
[email protected]), University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701. e BethToby GrossmanDistinguishedProfessor inSpirituality andHealing, (
[email protected]); AssociateRegional Dean, UTSPHAustinRegional Campus, 1616Guadalupe Street, Suite 6.300, Austin, TX 78701; Professor, Division of Epidemiology, Human Genetics, and Environmental Sciences, University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701.
Address correspondence to: Kathleen R. Case, Postdoctoral Trainee, (
[email protected]), University of Texas School of Public Health, Austin Regional Campus, 1616 Guadalupe Street, Suite 6.300, Austin, TX 78701. This study was partially funded by the Michael & Susan Dell Foundation through resources provided at the Michael & Susan Dell Center for Healthy Living, The University of Texas School of Public Health, Austin Regional Campus.
832 •
Journal of School Health
•
November 2016, Vol. 86, No. 11
•
© 2016, American School Health Association
graders reported ever smoking in 20134 and overall reductions in youth cigarette smoking prevalence have slowed in recent years.3 Predictors of Youth Tobacco Use To develop effective interventions to reduce the prevalence of tobacco use, research has been conducted to determine the predictors of tobacco use in youth. Personal and social variables are among the strongest predictors, such as sensation seeking,6 depression and anxiety,3,7 peer smoking,8 and parent smoking.9 Furthermore, lesbian, gay bisexual, and transgender youth are also at an increased risk of using tobacco products as compared with heterosexual youth.10 In addition to psychosocial factors associated with adolescent tobacco use, one of the strongest predictors of cigarette smoking is peers’ smoking habits.11 Although previous research has focused largely on social norms around smoking and the smoking behaviors of friends, less research has been devoted to examining the influence of negative peer relationships, specifically bullying, on tobacco use behaviors in youth.
Bullying Bullying is highly prevalent among US middle and high school students. National estimates indicate that 20.1% of high school students reported at least one instance of being bullied during the past year at school.12 While many studies that assess bullying focus on in-person bullying, such as teasing and/or physical intimidation, the ubiquitous nature of cellphones and the widespread use of social media sites such as Facebook have introduced a new form of peer victimization: electronic bullying.13-15 Electronic bullying, also referred to as ‘‘cyberbullying’’ is defined by Tokunaga as ‘‘any behavior performed through electronic or digital media by individuals or groups that repeatedly communicates hostile or aggressive messages intended to inflict harm or discomfort on others.’’14(p278) In a study that examined the prevalence of bullying in the past 2 months in a sample of middle school youth, 13% of students reported being the victim of physical bullying, 37% reported being the victim of verbal bullying, and 10% reported being a victim of electronic/cyberbullying.16 The prevalence of bullying differ according to sex and grade.17,18 For example, males are generally more likely to experience overt physical bullying, while females are generally more likely to experience bullying in the form of relational victimization, such as being the target of rumors, social isolation, or shunning.17 Bullying also varies by age, with middle school students reporting bullying more often than high school students.18 Journal of School Health
•
Bullying and Substance Use Previous research suggests that the association between bullying and substance use is complex; with some research indicating that the association differs according to bullied status and grade level.19 In one study, researchers concluded that adolescents who were both bully perpetrators and victims of bullying (‘‘bully-victims’’) reported the highest prevalence of substance use; there was a consistent positive association between being a bully-victim and substance use for both middle- and high-school students across substance type (marijuana, alcohol, and cigarettes).19 The association between being a victim of bullying and substance use was significant only for high school students; with victims of bullying reporting a higher prevalence of cigarette smoking and alcohol use compared with their nonbullied peers.19 TharpTaylor et al20 determined that sixth to eighth grade youth who reported being a victim of bullying were significantly more likely to report using cigarettes than their nonbullied peers. One study found that adolescents who reported being a victim of electronic bullying in the past year had significantly higher odds of high substance use, defined as using 3 or more substances (tobacco, alcohol, marijuana, inhalant, and other drugs), compared with their nonbullied peers.21 Furthermore, in one of the only studies to examine the association between cyberbullying and cigarette use, researchers found that in a sample of 6th, 8th, and 10th graders victims of cyberbullying were more likely to report smoking cigarettes than their nonbullied peers.22 To date, however, no studies have been conducted examining associations between being bullied at school only, electronically only, and both at school and electronically, and tobacco use behaviors in US high school students. Theoretical Rationale One potential explanation for the association between bullying and tobacco use is the theory of self-medication, which states that individuals who experience depression or anxiety may use substances in an effort to reduce symptoms of stress.23,24 This rationale is supported by previous literature which demonstrates that adolescents who are the victim of bullying are at increased odds of suffering from depression and anxiety.25,26 Furthermore, symptoms of depression and anxiety in adolescents are associated with increased odds of cigarette smoking. 7,27,28 Ultimately, adolescents who are victims of bullying may experience symptoms of depression and anxiety, and, therefore, may use tobacco products to alleviate such symptoms. Study Purpose This study addresses important gaps in the literature regarding the association between bullying and
November 2016, Vol. 86, No. 11
•
© 2016, American School Health Association • 833
tobacco use in adolescents. First, this study explores the specific association between being bullied and tobacco use; most of the previous research has examined the relationship between bullying and substance use generally (tobacco, alcohol, and illicit drugs),18,29,30 with only a few studies that have previously explored the association between bullying and cigarette smoking in youth.19,20,22 Next, this study examines the associations between being bullied at school only, being bullied electronically only, and being bullied both at school and electronically (‘‘bullying source’’) and tobacco use. It is hypothesized that for both sexes, adolescents who are victims of bullying will have higher odds of ever use of cigarettes, current use of cigarettes, and current use of any tobacco products as compared with adolescents who were not bullied. Second, it is also hypothesized that adolescents who experience bullying at school and electronically will have the highest odds of ever use of cigarettes, current use of cigarettes, and current use of any tobacco products compared with adolescents who experience only one type of bullying such as only at school or electronically.
METHODS Study Design This study was based on a cross-sectional analysis using nationally representative data of 9th to 12th grade students from the national 2013 Youth Risk Behavior Survey (YRBS). The YRBS is a surveillance system conducted by the CDC on a wide range of adolescent health-risk behaviors, including: tobacco use, alcohol and other drug use, physical inactivity, violence, among others.31 The YRBS is conducted nationally and in participating states; the national survey is conducted biennially in odd years.31 Previous research has documented the test-retest reliability of the YRBS; the questions have an average kappa of 0.61, corresponding to moderate reliability.32 Participants The sample for the 2013 YRBS included 13,583 participants; of the nonmissing data 6621 were girls, 6950 were boys, corresponding to a weighted percentage of 50% girls and 50% boys. With respect to grade, there were 3588 9th graders, 3152 10th graders, 3184 11th graders, and 3557 12th graders. For the purposes of this study, race was collapsed into 4 categories: white, black, Hispanic, and other race. Of the study participants, 5449 were white non-Hispanic, 2993 were black, 3395 were Hispanic, and 1428 were classified as other race. Measures Tobacco use. Three tobacco use behaviors served as the dependent variables for the analyses: ever use of cigarettes, current use of cigarettes, and current use of 834 •
Journal of School Health
•
November 2016, Vol. 86, No. 11
any tobacco product. The ever use question from the YRBS asked, ‘‘Have you ever tried cigarette smoking, even one or two puffs?’’ with response options yes and no.33 To determine current cigarette use, responses to the following question were analyzed: ‘‘During the past 30 days, on how many days did you smoke cigarettes?’’ For the purposes of our analyses responses were collapsed into a dichotomous variable, with any response equal to or greater than 1 or 2 days coded as ‘‘current user’’ and a response of 0 days coded as ‘‘not current user.’’ Current use of any tobacco products was operationalized as past 30-day use of cigarettes, chew/smokeless tobacco, and cigars/cigarillos, using the same criteria as current cigarette use. Bullying source. To determine bullying source, responses to 2 YRBS questions were analyzed. The 2013 YRBS defined bullying as, ‘‘Bullying is when 1 or more students tease, threaten, spread rumors about, hit, shove, or hurt another student over and over again. It is not bullying when 2 students of about the same strength or power argue or fight or tease each other in a friendly way.’’33 For bullying at school the question included, ‘‘During the past 12 months, have you ever been bullied on school property?’’ For electronic bullying the question included, ‘‘During the past 12 months, have you ever been electronically bullied? (Include being bullied through e-mail, chat rooms, instant messaging, Web sites, or texting).’’ The responses for both questions were yes and no. From these two questions 4 categories of bullying source were developed: not bullied, bullied at school only, bullied electronically only, and bullied at school and electronically. First, participants who responded no to both being bullied at school and bullied electronically were classified as not bullied. Participants who responded yes to being bullied at school but no to being bullied electronically were grouped in the bullied at school only category; participants who responded no to being bullied at school and yes to being electronically bullied were included in the bullied electronically only category, and students who responded yes to both questions were grouped in the bullied at school and electronically category. Additional covariates. Covariates included in the analyses were sex, age, race/ethnicity, weight status, depressive symptoms, and alcohol use. Demographic factors were based on single-item self-reported measures and included sex, age, and race/ethnicity. Sex was operationalized as a dichotomous variable. Age was included as a continuous variable with values from 12 to 18 years. A total of 3 race categories were created for the analyses: Hispanic/Latino, black, and other race. White non-Hispanic adolescents served as the referent category for the logistic regression analyses. Given research that has found bullying to be associated with weight status,34,35 students’ weight status was added as a covariate in the analyses. For •
© 2016, American School Health Association
this study, normal weight was classified as a body mass index (BMI) less than the 85th percentile for age and sex; overweight/obese status was defined as those adolescents with a BMI at or above the 85th percentile for age and sex.36,37 Normal weight served as the referent category for analysis. Depressive symptoms and current alcohol use were included as potential confounders, as previous research has documented that both are associated with tobacco use in adolescents.3,25,38 Depressive symptoms were operationalized using a dichotomous variable; response options were yes and no. For the purposes of analyses, adolescents who indicated that they did not experience depressive symptoms during the past 12 months served as the referent group. Current alcohol use was included as covariates as a dichotomous variable; adolescents who reported not using alcohol during the past 30 days served as the referent category for current use. Data Analysis All statistical analyses were conducted using STATA 12.0 (College Station, TX). Chi-square analyses were conducted to determine if there were significant differences in the prevalence of ever cigarette use, current cigarette use, and current tobacco use across sex, grade, weight status, and bullying source variables. The prevalence of being bullied across different bullying sources was calculated across sex, grade, and race/ethnicity variables. Chi-square tests also were conducted to determine if there were significant differences in the prevalence of source of bullying across demographic variables. Multivariable logistic regression analyses using national sampling weights were employed to determine the association between bullying source, and ever cigarette use, current cigarette use, and current tobacco use, adjusting for age, race/ethnicity, weight status, depressive symptoms, and alcohol use, stratifying by sex.
RESULTS The prevalence of ever cigarette use, current cigarette use, and current use of any tobacco products by sex, grade, race/ethnicity and bullying source are presented in Table 1. There were significant differences in the prevalence of ever smoking by sex, grade, race/ethnicity, and bullying source (p < .05). For results by bullying source, adolescents who reported being bullied both at school and electronically reported the highest prevalence of ever cigarette use (56.8%). For current cigarette use, there were significant differences by grade, race/ethnicity, and bullying source (p < .05). Approximately 1 in 4 adolescents (25.1%) who reported being bullied both at school and electronically reported current cigarette use compared Journal of School Health
•
with 14.3% of students who reported not being bullied in the past 12 months. For current use of any tobacco product, there were significant differences by sex, grade, race/ethnicity, and bullying source. Adolescents who were bullied both at school and electronically reported the highest prevalence of current use of any tobacco product (33.5%). Results for the logistic regression analyses are presented in Tables 2 and 3. The unadjusted odds ratios indicate that, compared with their nonbullied peers, adolescents who were bullied electronically only and both at school and electronically had significantly higher odds of ever use of cigarettes, current use of cigarettes, and current use of any tobacco products. In addition, girls who were bullied at school only also reported higher odds of ever use of cigarettes compared with girls who were not bullied. After adjusting for covariates, girls who were bullied both at school and electronically had significantly higher odds of ever smoking (adjusted odds ratio [ORadj] = 1.88, 95% confidence interval [CI] = 1.32-2.61) compared with girls who were not bullied. The results for girls who were bullied at school only approached significance (p = .05), with girls who were bullied at school only having higher odds of ever use compared with girls who were not bullied (ORadj = 1.27, 95% CI = 1.00-1.62). For current use of cigarettes, girls who were bullied both at school and electronically had significantly higher odds of use compared with girls who were not bullied (ORadj = 1.54, 95% CI = 1.032.32). For current use of any tobacco product, girls who were bullied electronically only and both at school and electronically had significantly higher odds of use compared with girls who were not bullied (ORadj = 1.44, 95% CI = 1.01-2.06; ORadj = 1.46, 95% CI = 1.06-2.00, respectively). For boys, after adjusting for covariates, only the association between being bullied electronically only and ever use of cigarettes remained significant; boys who were bullied electronically only had significantly higher odds of ever smoking compared with boys who were not bullied (ORadj = 1.78, 95% CI = 1.08-2.94).
DISCUSSION The purpose of this study was to determine the association between tobacco use and bullying source, specifically examining if there were differences between adolescents who were bullied at school only versus those who were bullied electronically only or both at school and electronically. In accordance with our first hypothesis, we found significant differences in the prevalence of tobacco use by bullying source; adolescents who were bullied both at school and electronically had the highest prevalence of ever use of cigarettes, current use of cigarettes, and current use of any tobacco product, as compared with adolescents
November 2016, Vol. 86, No. 11
•
© 2016, American School Health Association • 835
Table 1. Prevalence of Tobacco Use (%) by Sex, Grade, Race/Ethnicity, and Bullying Source, Using Sampling Weights; 2013 YRBS
Variable Sex Girls (N= 6621) Boys (N= 6950) Total (N= 13,571) p-Value Grade 9th (N= 3588) 10th (N= 3152) 11th (N= 3184) 12th (N= 3557) p-Value Race/ethnicity White, non-Hispanic (N= 5449) Black (N= 2993) Hispanic (N= 3395) Other race (N= 1428) p-Value Bullying source Not bullied (N= 10,256) School only (N= 1355) Electronically only (N= 732) School and electronically (N= 1144) p-Value
Ever Use of Cigarettes % Yes (95% CI) (N = 5360)
Current Use of Cigarettes % Yes (95% CI) (N = 1888)
Current Use of Any Tobacco % Yes (95% CI) (N = 2813)
39.7 (36.3-43.1) 42.5 (39.9-45.2) 41.1 (38.4-43.8) .04
14.9 (12.5-17.9) 16.4 (14.3-18.7) 15.7 (13.5-18.1) .13
17.7 (15.2-20.5) 27.0 (24.4-29.7) 22.3 (20.0-24.9)
