Publication Date




Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Psychology (Arts and Sciences)

Date of Defense


First Committee Member

Alan M. Delamater

Second Committee Member

Annette M. La Greca

Third Committee Member

Patrice G. Saab

Fourth Committee Member

Anna Maria Patino-Fernandez

Fifth Committee Member

Kristin M. Lindahl


Objectives: Prior research indicates that minority youth are more at risk for developing obesity, and that obesity is associated with several health complications as well as significant healthcare costs. The current study addressed the role of symptoms of internalizing disorders with regards to obesity in primarily ethnic minority fourth and fifth graders. In addition, this study was the first to expand upon prior research by determining whether symptoms of depression, anxiety, and social anxiety were associated with BMI at six month follow up in this age group, and whether gender moderated the relationship of internalizing symptoms and BMI. Finally, the current study also determined whether physical activity and diet mediated the relationship between depression, anxiety, social anxiety and BMI. Method: This study had a two-wave prospective design, with baseline and 6 month follow up data collection. Ninety-three children were recruited in fourth (n = 53, 56%) and fifth (n = 40, 43%) grades at two public elementary schools. Research staff recruited children via letters sent home to the parents. Parents completed consent forms and demographic data. Research staff collected height, weight, and questionnaire data from youth during the spring or fall of the academic year, dependent upon school, at baseline (May 2011 for Time 1 school 1; November 2011 for Time 1 school 2) and collected height and weight at six month follow up (November 2011 for Time 2 school 1; May 2012 for Time 2 school 2). The sample was 46% girls and youth were from predominantly ethnic minority backgrounds (76% Hispanic). At Time 1, 28% (n = 26) of children were considered overweight and 19% (n = 18) were obese. At time 2, 33% (n = 31) of children were considered overweight and 25% (n = 23) were obese. Questionnaires included the Children’s Depression Inventory- Short Form; Revised Children’s Manifest Anxiety Scale, Second Edition Short Form; the Social Anxiety Scale for Children-Revised; several diet and physical activity items from the Youth Risk Behavior Survey; and the Pubertal Development Scale. Results: Hierarchical regressions examined whether: 1) symptoms of depression, anxiety, and social anxiety were concurrently associated with BMI; 2) depressive, anxious, and socially anxious symptoms predicted later BMI at time 2; 3) internalizing symptoms predicted a change in BMI z-scores; 4) internalizing symptoms predicted overweight/obese status; 5) physical activity and diet mediated these relationships; and 6) gender moderated the prospective relationship between internalizing symptoms and BMI. Findings revealed that the shared variance between internalizing symptoms was associated with BMI at Time 1, R2change= .08; Fchange (3, 86) = 3.06, p = .03, but no internalizing symptoms uniquely predicted BMI. The predictive relationships between symptoms of depression and anxiety and BMI at Time 2 were not significant, nor were the relationships between internalizing symptoms and change in BMI or likelihood of being overweight or obese. The mediation analyses were not able to be adequately tested due to lack of significant findings in the prospective analyses and because the potential mediators were not related to children’s BMI. Finally, there was a significant interaction of gender on the relationship between social anxiety and BMI at Time 2, R2change= .06; Fchange (3, 78) = 4.75, p = .004. Further examination of simple slopes illustrated that, for girls only, social anxiety symptoms were predictive of BMI at Time 2, SASCR ß = .35, p = .001, such that higher social anxiety symptoms were associated with higher BMI at Time 2. Discussion: These results extend prior research in this field by demonstrating that, as a whole, internalized distress or negative affect is related to children’s BMI, as it was the shared variance between depression, anxiety, and social anxiety that was associated with children’s BMI. Furthermore, findings from the prospective analyses revealed relationships between social anxiety and BMI for girls in a primarily Hispanic sample of children. Further exploration of the relationships between depression, social anxiety, and general anxiety, or general negative affect, and BMI in minority children is needed to determine the nature of the prospective relationships (i.e., whether BMI predicts later internalizing symptoms or internalizing symptoms predict to later BMI). Also, an examination of potential underlying mechanisms (e.g., physical activity) would further inform these relationships and assist in identification of potential targets for intervention. Varied measurement strategies (e.g., daily recalls, interviews, accelerometry) for evaluating these underlying health behaviors (i.e., diet, physical activity) are needed to better evaluate the potential contributing mechanisms. Current findings underscore the importance of utilizing screeners of internalizing symptoms or general negative affect to identify children who may be at risk for becoming overweight. Also, findings highlight the importance of measuring social anxiety as a potential means of identifying girls who may be at risk for weight concerns or for developing obesity. Efforts to reduce internal distress in youth, and social anxiety specifically in girls, may have a positive health benefit and requires further exploration.


obesity; depression; anxiety; children; childhood obesity