Adolescent obesity is an area of growing public health concern. The Centers for Disease Control and Prevention conducts surveys through their Youth Risk Behavior Surveillance System (YRBSS) every two years to monitor a variety of health risk factors and behaviors among high school and middle school students. The YRBSS compiles information about obesity and BMI percentile, among many other factors. We accessed a combined dataset available on the YRBSS website which includes all data collected from high school students’ surveys from 1991 to 2013. Due to updating of questionnaires and adding of variables over the years, some variables only appear in the most recent years, limiting trend analysis to the timeframe in which the variable of interest was included. We analyzed the linear and quadratic trends in BMI percentile, obesity, and overweight rates in the national Youth Risk Behavior Survey (YRBS) from 1999 to 2013 and in the Tennessee YRBS from 2003 to 2013. Each variable was stratified by age and race to observe differences among groups. National trends show an increase in average BMI percentiles overall from 1999 to 2007, a decrease between 2007 and 2009, then another increase between 2009 and 2013. Tennessee trends show a dramatic increase in average BMI percentile overall from 2003 to 2007, then a decrease between 2007 and 2013. Since 2005, Tennessee has maintained a higher average BMI percentile (64.23, 66.37, 65.00, 64.96, 64.23) than the national average (63.47, 64.23, 62.81, 63.00, 63.51) for each recorded year, however, the decreasing trends in Tennessee and increasing trends in the nation have brought the average BMI percentiles of each to comparable rates. There is literature to support the variation of BMI among young, middle-aged, and elderly individuals. However, there is currently little evidence of differences in BMI percentiles, obesity, or overweight rates between different age groups of high school students. We expect to see little, if any, differences across different age groups of high school students in this study both nationally and at the state level. Racial and ethnic disparities exist for a variety of health conditions and outcomes. Many conditions, including obesity, disproportionately affect minority populations. We expect to see differences in BMI percentiles, obesity, and overweight rates across different races at both nationally and at the state level.

Disease mapping is used to identify high risk areas, inform resource allocation and generate hypotheses. The stroke and diabetes belts in the U.S. have encouraged public dialogue and spurred research. Lung cancer is the leading cause of U.S. cancer mortality, accounting for 158135 deaths in 2010 compared to 129180 from cerebrovascular disease and 68905 from diabetes mellitus. If one exists, defining a distinct pattern of high lung cancer mortality could increase public awareness of the disease and facilitate investigation of its determinants. To begin our inquiry, we generated a map and observed an area of high lung cancer mortality, primarily in the Southeast. However, variability in county rates, likely due to small populations, made determining patterns difficult. Spatial smoothing can clarify obscured patterns. We downloaded county lung cancer mortality rates, population sizes and death counts. Concurrent incidence and mortality rates for lung cancer were nearly equivalent, so mortality was used as a proxy for risk. After downloading county population centroids with latitudes and longitudes, we implemented a median-based, weighted, two-dimensional smoothing algorithm to enhance spatial patterns by borrowing strength from neighbor counties. The algorithm selected three proximate centroids, forming a “triple,” anchored by the centroid of the county to be smoothed. The parameter for nearest neighbor (NN) counties was set to NN=10, with the number of triples (NTR) for each county NTR=(2/3)*NN, producing seven collinear triples for each county with a center angle ≥135°. Median rates for the top and bottom 50% of neighbor counties were calculated and weighted by 1/SE, creating a “window,” whereby if the original rate was between the two medians, or if the county population was sufficiently large, it was not smoothed. If the original rate was outside the window, it was adjusted according to the corresponding neighbor median. Ten iterations of this process were conducted for each county. Smoothed rates were imported to ArcGIS and joined to a U.S. counties layer. Congruent counties in or near the Southeast with rates above 64 per 100,000 were defined as one class. We observed clustering of high lung cancer mortality, comprising 724 counties and forming an arc not evident in the unsmoothed data. This area, which we define as the lung cancer belt, included nearly all of Arkansas, Kentucky and Tennessee, and portions of 16 other states. Heavily affected regions include much of the Ohio Valley, Central Appalachia, the Tennessee Valley, the Ozarks, the Mississippi Delta and the northern Gulf Coast. Smoking, a modifiable behavior, causes the majority of lung cancer deaths, and is the single leading cause of mortality in the United States. Lung cancer mortality rates presented at the state level obscure differences within states. The lung cancer belt may provide a tool to identify areas in greatest need of resources. National survey data could be utilized to determine demographic, socioeconomic and behavioral differences between the lung cancer belt and the rest of the nation.

Background: More than one-third of adults and 17% of children/adolescents in the US are overweight or obese contributing to significant morbidity and mortality, and healthcare costs. Little is known about obesity patterns in subethnic populations in the US with high prevalence of obesity or overweight reported in Hispanic youth (21.9%) compared to non-Hispanic blacks (19.5%) and non-Hispanic whites (14.7%). The purpose of this study is to assess the association of generation status with adolescent overweight or obesity in Hispanics in the US. Data from the 2011-2012 National Survey of Children’s Health were used to conduct this study, focusing on Hispanic adolescents aged 10 to 17 years. Adolescents were defined as being overweight or obese if they were above 85th or 95th percentile cut-off point values for BMI. Conclusion:Generation status of an adolescent was categorized into three groups depending on the adolescent and the parents’ place of birth. Multivariable models were conducted to test the association of generation status with adolescent obesity in Hispanics, adjusting for potential confounders. Comparing to 3rd generation adolescents, those belonging to 1st and 2nd generations were associated with increased odds of adolescent obesity in Hispanics OR 1.48, 95% CI 1.177 – 1.867 and OR 1.405, 95% CI 1.227-1.610 for 1st and 2nd generation, respectively. Generational status is associated with increased relative odds of overweight or obesity in Hispanic adolescents. Aggregated estimates not accounting for country of origin of an adolescent contribute to significant heterogeneity or disparities in obesity prevalence or patterns, with implications for generation-specific interventions

Background: Vitamin D is known to be associated with inflammatory diseases, but its relationship with allergic diseases is unclear. The study objective is to determine the association of serum vitamin D levels and markers of wheeze, asthma and atopy. Methods: Data (n = 9,463) on serum vitamin D levels and atopy were obtained from 2005-2006 National Health and Nutrition Examination Survey. Serum vitamin D level was categorized into four groups: Normal (≥30ng/ml), Insufficient (21-29ng/ml), Deficient (11-20ng/ml) and Severely Deficient (≤10ng/ml). Atopy was defined as at least 1 positive allergen-specific IgE level measured for a panel of 5 common aeroallergens- cat, dog, house dust mite, cock roach and Alternaria species. Doctor-diagnosed asthma and wheeze in the previous 12 months were assessed by means of questionnaire. Multivariable logistic regression analyses were conducted to investigate the association of serum vitamin D with wheeze, asthma and atopy adjusting for age, sex, race, smoking, outdoor physical activity, body mass index and poverty income ratio. Results: Overall, 15%, 14% and 28% of subjects had wheeze, asthma and atopy, respectively. Approximately 21% had normal serum vitamin D levels, while 35%, 28% and 5% had insufficient, deficient and severely deficient levels. Compared to subjects with normal vitamin D levels, those with insufficient, deficient and severely deficient levels had increased relative odds of wheeze and atopy with highest adjusted estimates in subjects with severe vitamin D deficiency (adjusted odds ratio [OR] 2.31, 95% Confidence Interval [CI] 1.73-3.10 for wheeze; OR 1.49, 95% CI 1.17-1.89 for atopy). Conclusion: Low serum vitamin D levels were found to be associated with wheeze and atopy. Findings contribute to ongoing efforts to understanding the role of vitamin D in atopic diseases.

Introduction. Suicide-related fatalities are the third leading cause of death among adolescents, resulting in approximately 4,600 deaths annually. According to findings from the 2005 National College Health Assessment, engagement in weekly physical activity (PA) reduced the risk of suicidal behaviors. Data from the 2010 middle school Youth Risk Behavior Survey (YRBS) administered by Tennessee Coordinated School Health this study investigates the cross-sectional relationships of PA with suicidal behaviors, while simultaneously considering explanatory variables such as personal characteristics, sedentary behaviors, drug use, extreme weight control behaviors (EWCB), body mass index (BMI) and weight misperception. Methods. This is a secondary analysis of data from the 2010 Tennessee Middle School YRBS conducted among 65,182 middle school students of which 60,715 students were included in the final analysis. Items assessed were PA, sports team engagement, PE class, sedentary behaviors, suicidal behaviors, drug use, EWCB, BMI, weight misperception and selected personal characteristics. A bivariate analysis was used to create simple descriptive statistics, including means, standard deviations and proportions. Overall, approximately 21.15% (N= 13,704) reported suicidal behaviors, specifically, 18.30% reported ever having thoughts of suicide, 11.13% had ever made a suicide plan and 7.02% had one or more suicide attempts in their lifetime. A logistic regression analysis was conducted to compare the odds of experiencing suicidal behaviors in adolescents who engaged in PA, sports teams, and PE class, with students who did not perform these activities. We used odds ratios (OR) to assess effect size (ES), rather than p-values to assess statistical significance due to the study’s large sample size. As general rules of thumb, there is a small effect size when ES=0.20 or OR = 1.44, 0.694, medium effect size when ES=0.50 or OR = 2.47, 0.405 and large effect size when ES=0.80 or OR = 4.25, 0.235. Results. Based on ES, age, race/ethnicity, gender, grade in school, sports team engagement, PE class attendance, sedentary behaviors, drug use and EWCB were associated with suicidal behaviors. Associations were strongest for drug/substance use and EWCB and suicidal behaviors, both with large ESs. Although weakened when controlling for other independent variables, the ES of race/ethnicity (other vs white non-Hispanic), gender, sports team engagement, drug/substance use and EWCB on suicidal behaviors were >0.20. The ES of age, PE class attendance, and sedentary behaviors were <0.20. Conclusion. Findings suggest that sports team engagement is associated with reduced risk for suicidal behaviors; whereas, no effects were found for PA or PE class attendance. Asking adolescents questions about sports team engagement may help screen for risk of suicidal behaviors.

Abstract available through http://www.fasebj.org/doi/abs/10.1096/fasebj.31.1_supplement.1037.5.

In this study, we used a cross-sectional study design to examine the relationship between the calcium intake and risk factors for CVD among obese adults by using continuous waves of National Health and Nutrition Examination Survey (NHANES) data 1999-2010. The association between calcium intake and risk factors of CVD (hypertension, total cholesterol, HDL, glycohemoglobin), CRP, albuminuria) is assessed among obese adults in USA. The incidence of Cardiovascular Disease (CVD) is high among obese people. The potential effects of inadequate calcium intake on CVD are receiving increased epidemiologic attention. Understanding the association between risk factors for CVD and calcium intake among obese adults is important for the advancement of CVD, nutrition and obesity research. Data collected from the National Health and Nutrition Examination Survey from 1999-2010 were examined, utilizing a subset of 14,856 obese subjects. Analysis of Variance statistical tests were conducted to determine associations between calcium intake and CVD risk. Simple and multiple linear and logistic regression analyses were conducted to determine the predicted value of calcium intake with CVD. After adjusting for energy intake and other potential risk factors, systolic blood pressure, diastolic blood pressure, C-reactive protein, glycosylated hemoglobin and albuminuria were negatively associated with calcium intake at &alpha = 0.05 level in both linear and logistic regression analyses. In a comparison of calcium intake by quartiles, results reveal that total cholesterol had a weak negative association with calcium intake at &alpha = 0.1 level. The implications of these study results are important as the importance of adequate calcium intake and its potential to decrease CVD among obese adults has incredible preventive value for populations. Additional research that focuses on dietary intake, calcium thresholds and impacts on total cholesterol levels in the body is warranted.

Introduction. Among U.S. adolescents (12-17 years), tobacco, alcohol and marijuana are commonly used substances. Recent data suggests that although trends in tobacco use have declined, marijuana and alcohol use rates are steadily increasing, especially among younger adolescents. The main purpose of this study was to characterize differences in tobacco, alcohol and marijuana onset among Appalachian and nonAppalachian middle school students in Tennessee. At present, there is limited research on tobacco and drug use among younger adolescents in Tennessee and Appalachia as a whole. Therefore, these findings have important implications for the assessment and prevention of risk behaviors among adolescents regionally and may help to establish priorities for policy and preventive measures. Methods. This is a secondary data analysis of the Youth Risk Behavior Survey (YRBS), consisting of data collected in 2010 from a representative sample (n=65,182) of middle schoolers in Tennessee. All analyses were performed on weighted data so that results represent all middle school students in Tennessee. Primary outcome variables included early age of onset (age) of alcohol, tobacco, and marijuana use. The primary exposure variable was region (Appalachia versus non-Appalachia). Multiple logistic regression was used to determine the association between early age of onset of cigarette, alcohol, and marijuana use, and region controlling for personal characteristics, other substance use, suicidal behaviors, body mass index, weight misperception and extreme weight control behaviors. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were reported. Results. Early initiators of cigarette, alcohol or marijuana had a two to four fold increase risk for experimentation with other substances compared to late age of onset, with cigarette use posing the highest risk [OR: 4.73, C.I. (4.47, 5.01)]. Middle schoolers in Appalachia were at greater risk for early age of onset of cigarette [OR: 1.502, C.I. (1.421, 1.587)] and marijuana use [OR: 1.268, C.I. (1.169, 1,375)] compared to non-Appalachia middle schoolers, and 80.9% less likely to use alcohol prior to age 12 [OR: 0.809, C.I. (0.777, 0.843). Other differences were observed for risk of early onset of use and gender, race/ethnicity, age, other substance use and EWCB. Conclusion. Differences in the pattern of substance use were observed for middle schoolers in Appalachia and non-Appalachia Tennessee. Better understanding of these differences will help inform public health policy and practice targeting cigarette, alcohol and marijuana use in the region.

Introduction: Childhood obesity has more than quadrupled in the last 30 years, with the prevalence in adolescents aged 12-17 years increasing from 5% in 1980 to 21% in 2012. The purpose of this study was to estimate the extent to which childhood obesity in Tennessee is associated with between-context differences (districts, schools and classes) and to identify factors at the district, school, class, and individual level that influence the individual weight status among 64,790 Tennessee children and adolescents. Methods: Crosssectional data from the Youth Risk Behavior Survey (YRBS) conducted in Tennessee (2010) were used to conduct multilevel analyses that account for the nesting of students in classes, classes in schools and schools in districts. The outcome variable was childhood obesity (>95th percentile). Explanatory variables included district-level factors (the proportion of children wearing seat belts or helmets in district and the proportion of being asked to show proof of age), school-level factors (current tobacco use in school, and HIV/AIDS education in school), class-level factors (the average of smoking days in past 30 days and the proportion of ever having exercised to lose weight in class) and individual-level factors (state geographical regions, age, gender, grade, ever ridden in a car driven by someone who had been drinking alcohol, ever carried a weapon, made a plan to kill yourself, ever used or early onset use of tobacco, alcohol, marijuana, exercised to control weight, school day television time, days of physical education (PE) classes. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were reported. Results: Multilevel analyses indicate that 0.90%, 0.08%, and 0.45% of the variation in obesity is associated with class, school and district differences, respectively. Male middle schoolers were at greater risk for obesity [OR: 1.82, C.I. (1.75, 1.89)] compared to females. For every one year increase in age, the relative odds of obesity increased by 11% (OR 0.89, 95% CI 0.88-0.91). Students with worse grades were more likely to have obesity [OR: 1.33, C.I. (1.13, 1.56)]. Students who watched TV 3 hours or more per day were more likely to be obese [OR: 1.31, C.I. (1.23, 1.40)] compared to those who did less than 3 hours per day. Similarly students who ever tried cigarettes were more likely to be obese [OR: 2.15, C.I. (1.62, 2.85)] compared to those students who did not. Students who reported wearing seat belts [OR: 0.05, C.I. (0.02, 0.16)] were less likely to be obese. Conclusions: This study highlights a number of modifiable factors on multiple levels associated with child and adolescent obesity in the state of Tennessee. The results emphasize the importance of targeting programs beyond individual adolescent factors to the child’s classes, schools, and school districts, to reduce the prevalence of obesity among Tennessee adolescents.

Matrix variate skew elliptically contoured distributions generalize several classes of important distributions. This paper defines and explores matrix variate skew elliptically contoured distributions. In particular, we discuss two stochastic representations of the matrix variate skew elliptically contoured distributions.