Detailed contribution information
| Contribution title | 3008 - Listen, I am afraid of gaining weight! Testing a new anxiety-based etiological model for anorexia nervosa |
|---|---|
| Contribution code | PS02-55 (P) |
| Authors | |
| Form of presentation | Poster |
| Topic |
|
| Abstract |
Objectives Although anorexia nervosa (AN) and anxiety disorders share great comorbidity, little interest has been raised for the predictive power of anxiety. While prior works have shown that increased levels of anxiety reduce the chance of successful recovery, it is unclear in what way anxiety impacts the course of eating disordered behavior (ED). We argue that anxiety represents a crucial factor in the etiology of AN by biasing the focus of attention. Hence, we strive to I) assess the predictive value of trait and specific anxieties on ED symptoms and aim to II) test whether (induced) anxiety influences the course of AN through biased attention. Methods First, a questionnaire-based study was conducted in a non-clinical (NCC) sample (14-25 years old) and adolescent ED patients of the LWL-university hospital for child and adolescent psychiatry (Hamm, Germany). 354 participants (301 NCC, 53 with ED) took part in the present study. ED symptoms were assessed by means of the Eating Disorder Examination Questionnaire (EDEQ), anxiety with the State Trait Anxiety Inventory, Fear of Negative Evaluation Scale, Fear of Positive Evaluation Scale and the Social Appearance and Anxiety Scale. Results Significant correlations were found between trait anxiety and ED symptoms (EDEQ-subscales ranged from r = .393 to r =.576). Associations became as strong as r = .671 for social appearance anxieties and ED symptoms. Trait anxiety, specific anxieties and BMI regressed hierarchically on each of the EDEQ subscales revealed that trait anxiety, social appearance anxiety and fear of negative evaluation are key predictors for all EDEQ subscales (range from R2 = .273 - .494). In particular, social appearance anxieties had the greatest impact on EDEQ subscales (range ß = .318 - .482). In contrast, BMI was only predictive for weight and shape concerns and added only 1% in explained variance. Conclusion There is clear evidence for the predictive power of anxiety, and in particular for social anxieties, on ED symptoms. In a second step, the causal role of anxiety for AN shall be studied experimentally in a group of AN patients, depressed patients and healthy controls. We induce anxiety to test whether anxiety biases the focus of attention and whether biased attentional processes represent a mediating factor between trait anxiety and AN symptomatology. Questionnaire-based and experimental data shall be presented and discussed in the light of theoretical and practical implications. |