Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Psychological and Brain Sciences

Degree Program

Clinical Psychology, PhD

Committee Chair

Levinson, Cheri

Committee Co-Chair (if applicable)

Depue, Brendan

Committee Member

Depue, Brendan

Committee Member

Hindy, Nicholas

Committee Member

Bufferd, Sara

Committee Member

Frey, Laura

Author's Keywords

eating disorders; adolescence; anxiety; prospective; risk factors; comorbidity


Introduction: Eating disorders are associated with significant morbidity, psychiatric comorbidity, and impairment. Despite the detrimental outcomes associated with eating disorders, effective treatments for eating disorders are lacking. One factor that has impeded the identification of targets for intervention in eating disorders is the high comorbidity rate with other psychiatric disorders. Comorbidity models can inform treatments by showing which mechanisms are shared in comorbidity and which mechanisms are unique to specific disorders. Anxiety disorders are the most frequently co-occurring disorders with eating disorders, as 85% of individuals with eating disorders have a comorbid anxiety disorder. One mechanistic process in both anxiety disorders and eating disorders is learning. Shared cognitive-behavioral learning mechanisms in both anxiety and eating disorders include repetitive negative thinking and negative affect. Unique cognitive-behavioral learning mechanisms to eating disorders include thinness expectancies and habit formation around eating. In the current study, I investigated the cognitive-behavioral mechanisms related to learning in comorbid eating disorders-anxiety. Method: Seventy adolescents (13-15 years old) from the community completed self-report assessments at three time points: at the beginning of school year, the mid-point of school year (four-month follow-up), and the end of school year (nine-month follow-up). Adolescence may be a critical period for the development of comorbid eating disorder-anxiety symptomatology. First, a cross-sectional path model was constructed to test whether negative affect, repetitive negative thinking, thinness expectancies, and habit formation were related to anxiety symptoms and eating disorder symptoms. Then, a prospective autoregressive model across the three time points (baseline, four-months, nine-months) was constructed to assess a comorbidity model of anxiety and eating disorder symptoms over time, while adjusting for baseline levels of symptoms. Finally, a mediation model from negative affect to eating disorders through thinness expectancies was tested. Results: Cross-sectionally, negative affect was a shared cognitive-behavioral mechanism significantly associated with both anxiety and eating disorder symptoms, and thinness expectancies was a unique cognitive-behavioral mechanisms associated with eating disorder symptoms. Due to significant attrition, exploratory models including two time points (baseline and four-month follow-up) and excluding habits around eating were conducted. Prospectively, only baseline eating disorder symptoms predicted eating disorder symptoms at four-month follow-up. No mediational effects were found. A post-hoc moderation analysis showed that individuals higher in thinness expectancies were more likely to have higher eating disorder symptoms when also higher in negative affect, compared to lower negative affect. Conclusions: This study was a preliminary test of an integrated learning model and hopes to guide future research regarding shared and unique vulnerability factors for eating disorder comorbidity. These findings suggest that the strongest predictor of later eating disorder symptoms in adolescence is earlier eating disorder symptoms. Thus, there is an urgent need to study eating disorder symptoms and comorbidity at an even earlier age, perhaps during preadolescence, before puberty.