An examination of responses to laboratory stressors in individuals with bulimia nervosa compared to those without a history of eating disorders

Abstract

The association between perceived interpersonal stress and symptoms of bulimia nervosa (BN) has been well documented. In line with this association, recent research is focused on better understanding common factors of BN and social anxiety (SA; e.g., fear of negative evaluation, perfectionism) that might contribute to the maintenance of BN symptomatology. Thus far, the extent to which stress affects individuals with BN and co-occurring SA has not been extensively studied, despite the high comorbidity of these two disorders. Using an experimental paradigm, the present study explored the effects of different stressor-types on individuals with BN, as well as the potential impact of SA symptoms and related factors on stress response in this population. A total of 56 females participated in this study: 28 diagnosed with BN, and 28 controls without a history of eating disorders. All participants completed self-report measures and underwent three laboratory stressors: a modified version of the Trier Social Stress Test (a psychosocial stressor), a carbon dioxide challenge (a physiological stressor), and a body image stressor (an eating disorder specific stressor developed for use in this study). Subjective and cardiovascular responses were measured. Contrary to hypotheses, significant between-group differences in stressor response were not consistently revealed. An examination of Cohen’s d values, however, suggested that the magnitude of group differences represented practically significant effects. Notably, participants within both groups reported decreased levels of hunger and desire to eat in response to each stressor. Also inconsistent with hypotheses was that neither SA symptoms nor associated factors consistently predicted stress reactivity in either group. Moreover, in the BN group, similar magnitudes of response were revealed across the three stressors. Implications and suggestions for further research are discussed. Although binge eating was not behaviourally measured, the finding that different stressor-types did not lead directly to increased hunger and desire to eat is notable and consistent with the cognitive behavioural theory of BN. With continued study of SA-related factors in association with BN symptomatology-- particularly in the context of stressful events-- further insight into the maintenance of this disorder might be revealed for a subset of individuals with comorbid BN and SAD.