Date of Graduation


Document Type


Degree Type



School of Public Health



Committee Chair

Geri Dino

Committee Co-Chair

Lesley Cottrell

Committee Member

Traci Jarrett

Committee Member

Christa Lilly

Committee Member

Keith Zullig


Introduction: Research on the health status of United States' youths has shown that the main adolescent health threats result from behavioral choices (health risk behaviors) that place boys and girls at risk for leading causes of morbidity and mortality. The aims of this study were to examine the co-occurrence of health risk behaviors and determine patterns of the co-occurrence of risk behaviors that may differ based on sex (aim 1); the association of health risk behaviors with weight status (aim 2); and the association of tobacco use, poor dietary habits, physical inactivity, and sedentary behavior among adolescents using nationally representative data (aim 3).;Methods: This study uses cross-sectional data from the Youth Risk Behavior Surveillance System (YRBSS) conducted between the years 2012 and 2013. The sampling frame for this survey included all adolescents in Grades 9 to 12 who attended regular public, private, or Catholic schools in the 50 states and the District of Columbia. The main outcomes of interest for the first two studies included indicators of health risk behaviors including substance use, intentional and unintentional injuries, risky sexual behavior, mental health, bullying, unhealthy weight control, poor dietary habits, physical inactivity, and sedentary behavior. The main outcome for the third study was tobacco use. Latent class analysis was conducted to determine the co-occurrence of health risk behaviors among adolescents. Regression analyses were conducted to examine the associations between weight status and risky behaviors for the second study; and tobacco use, poor dietary habits, physical inactivity, and sedentary behavior.;Results: With respect to findings for aim one, the latent class analysis yielded five distinct subgroups (classes). The majority of adolescents belonged to the low risk subgroups; however, our analyses identified three distinct high risk subgroups, high risk-- depressed and suicidal youth, high risk- tobacco and alcohol users, and high risk-polysubstance users. Girls and boys both reported low risk and high risk behaviors. However, girls had a distinct subgroup with particularly high probabilities of depressive symptoms and suicidal behavior. The second aim examining the association of weight status and risky behavior found that compared to normal weight students, obese students were more likely to use cigarettes (AOR = 1.44, 95% CI, 1.15-1.79), less likely to engage in behaviors that result in unintentional injury (e.g. driving drunk, riding with a drunk driver, texting and driving) (AOR = 0.83; 95% CI, 0.73-0.95), more likely to get bullied on school property, and electronically (AOR = 1.47, 95% CI, 1.28-1.68; AOR = 1.20, 95% CI, 1.003-1.447, respectively) and more likely to engage in unhealthy weight control (AOR = 7.56, 95% CI, 6.33-9.04 ). Findings for the third aim showed that both male and female adolescents who engaged in poor dietary behavior were more likely to use tobacco than those who did not engage in poor dietary behavior. For instance, both boys and girls who skipped breakfast every day for the past seven days were more likely to use tobacco than those who did not skip breakfast (AOR=1.54, 95% CI 1.16-2.03 and AOR = 1.32, 95% CI 1.06-1.64, respectively). Physical inactivity and playing video or computer games more than three hours per day yielded an inverse association with tobacco use among boys (AOR = 0.62, 95% CI 0.49-0.79).;Conclusions: Distinct subgroups of risky behaviors were identified for the study sample. Differences were found in the pattern of the co-occurrence of risky behavior by sex. Obesity was found to be positively associated with most of the risk behaviors examined in this project. The results of the third aim found that tobacco use was significantly associated with poor dietary habits, physical inactivity, and sedentary behavior. Longitudinal studies of risky behavior among adolescents are needed to establish definitive conclusions and to prove causality. In addition, future research is needed to evaluate risk behaviors among adolescents by sociodemographic factors and geographic region.