(PS7-25) Domain-specific Sources of Stress and Eating Disorder Symptoms

Abstract Text: Life stress has been linked to eating disorder (ED) symptoms and is an explicit target in cognitive behavioral treatment for EDs. However, few studies have examined the relative effects on ED symptoms of stress emanating from different life domains, and even fewer have used non-self-report, interviewer-based assessments of life stress. In the present study we examined in a nonclinical sample the association between self-reports of ED symptoms and interviewer ratings of stress in the domains of friendships, social life, romance, family, academics, work, finances, health of self, and health of family. Participants (N=926; 53% women) were part of an ongoing longitudinal study of college alcohol use and well-being; data for the present study was obtained from the second assessment wave which occurred 2-3 years post college. During this wave participants completed the Eating Disorder Examination Questionnaire (EDE-Q) and a phone-based semi-structured life stress interview. Results indicated significant (ps < .05) bi-variate associations between the EDE-Q global score and academic stress (r=.088), family-related stress (r=.120), and personal health stress (r=.131). Only two stressors (academic and personal health) were unique predictors in the positive direction in a multiple regression model. Unexpectedly, social stress emerged as a significant negative predictor of the EDE-Q global score in the multiple regression model. Total stress (i.e., the average of all domains) was also significantly related to the EDE-Q global score (r = .100) as well as several of the EDE-Q subscales including: Eating Concern (r=.105), Shape Concern (r=.129), and Weight Concern (r=.114). Total stress was not significantly correlated with Restraint (r = .01), a subscale that primarily evaluates behavioral symptoms related to eating restriction. Our findings indicate that specific types of stress may be associated with ED symptoms, and specific types of ED symptoms may be more strongly associated with stress. However, the weak observed associations highlight the need to examine potential moderators of these associations. Understanding the specific stressor-symptom relationships involved in EDs, and who might be more vulnerable to the effects of life stress, could help tailor cognitive behavioral interventions to individual clients experiencing different types of stressors. Additionally, understanding the role of stress in cognitive versus behavioral ED symptoms could enable clinicians to utilize stress management techniques strategically, and encourage patients with ED cognitions to closely self-monitor their ongoing chronic stressors.