Understanding Effects of Exercise and Diet to Improve Mental and Physical Health in Children With Behavioral Health Disorders

Abstract

Approximately 13%–20% of children living in the United States experience a diagnosable behavioral health disorder (BHD) in a given year. Children with BHDs have elevated risk of poor mental and physical health outcomes. It is established that diet quality and physical activity (PA) levels are predictors of both chronic disease risk and behavioral health. Unfortunately, children with BHDs experience increased risk of poor diet and low PA relative to typically developing peers. Given the high prevalence of BHDs and associated health disparities, this dissertation aims to help address major gaps in existing research. Chapter one investigates the effects on behavior of an RCT using cybercycling at a therapeutic school serving children with BHDs. We used mixed-effects logistic regression to assess relationships between intervention exposures and behavioral outcomes. We found that students successfully engaged in and had significantly lower risk of poor behavioral outcomes during the intervention, particularly on days they participated in aerobic exercise. Chapter two examines exercise dose and changes in behavior, to better inform exercise prescription and interventions for children with a variety of BHDs. We used mixed-effects linear regression to examine dose-response relationships between aerobic exercise duration and intensity, and minutes of disciplinary time out of class and self-regulation scores. We found duration had an inverse, linear relationship with negative outcomes, and effects were amplified among children with an Attention Deficit-Hyperactivity Disorder (ADHD) diagnosis. Chapter three considers roles played by ADHD diagnosis, stimulant use, and health behaviors in predicting children’s BMI change. We applied linear regression to longitudinal data from a nationally-representative cohort of children to assess associations between duration of stimulant use, ADHD diagnosis, and BMI change, and logistic regression to assess odds of poor diet and PA patterns between groups. We found stimulant use predicted greater BMI change in pre-/early adolescence, and children with ADHD were at elevated risk of poor dietary patterns regardless of medication status. Diet and PA did not mediate relationships between stimulant use and BMI change. Each chapter provides important evidence to address chronic disease and behavioral health risks among children with BHDs in order to reduce existing health disparities.