Determining the potential link of self-compassion with eating pathology and body image among women : a longitudinal mediational study

PURPOSE: This longitudinal study aims to determine what factors mediate the previously established link between self-compassion and eating pathology/body image concerns, over a 6-month period. METHODS: A community sample of 274 adult women (M = 29.50 years) completed standardised validated measures of self-compassion (Self-Compassion Scale), rumination (Ruminative Thought Style Questionnaire), shame (Other as Shamer Scale), perfectionism (Short Form of the Revised Almost Perfect Scale), self-criticism (Levels of Self-Criticism Scale), eating pathology (Eating Disorder Examination Questionnaire) and body image (Body Shape Questionnaire). They reported levels of: self-compassion at Time 1, potential mediators (rumination, shame, self-criticism, perfectionism) at 3 months; and eating pathology and body dissatisfaction a further 3 months later. Missing data were handled using multiple imputation. Stepwise multiple regression showed that shame was the most consistent mediator. RESULTS: Shame acted as a full mediator of the self-compassion-eating/body image relationship {respectively, [B = .04, SE = .01, t(268) = 3.93, p < .001], [B = .33, SE = .15, t(268) = 2.25, p < .05]}. Discrepancy perfectionism also played a mediating role in the link between self-compassion and body image dissatisfaction [B = .59, SE = .28, t(268) = 2.10, p < .05]. CONCLUSION: These results support the hypothesis that self-compassion is relevant to eating pathology and body image disturbance, and demonstrate that shame is an important mechanism in that relationship. This pattern suggests that interventions that reduce shame should be considered when addressing issues relating to self-compassion and its links to eating disorders. LEVEL OF EVIDENCE: Level IV, multiple time series without intervention.