Attachment representations in community girls at risk of Binge Eating Disorder.

Introduction. Among community adolescents, binge-eating is the prevalent symptom of Eating Disorder (ED), surpassing restrictive or compensatory behaviors (Marzilli, Cerniglia and Cimino, 2018). Findings in the attachment research reported a greater risk of ED’ symptoms for teenagers with insecure or disorganized attachment representations (Gander, Sevecke and Buchheim, 2015), but scarce studies investigated the relations between binge-eating symptoms and attachment in normative adolescents, usually using self-report questionnaires (Laghi et al., 2012). Therefore, in this comparative study we applied a semi-structured interview to compare attachment representations in community girls at risk or not for binge eating symptoms. Methods. Into a larger research, 382 high-school students (Mage =15.6y, 38% boys) were assessed in their risk for binge-eating symptoms, using the well-known specific self-report Binge Eating Scale (BES; Gormally et al., 1982), that measures the levels of binge-eating symptoms (at-risk with scores>17). For this study, we sub-sampled as participants 44 community girls (aged 14-18y, M=15.7, SD=1.1), 22 at-risk for binge-eating symptoms (Binge-Eating, BE) and 22 not-at-risk (NotBinge-Eating, NBE), preliminary compared as significantly different in BES scores (p=.000). Measures were: 1) the BES, used in the screening phase; 2) the Friends and Family Interview (FFI, Steele and Steele, 2005), a semi-structured interview to assess adolescents’ attachment representations as Secure-Autonomous, Insecure-Dismissing, Insecure-Preoccupied, Disorganized-Disoriented, both in terms of classifications and scores, focusing on narrative’s coherence. Results. There were no significant differences between groups in FFI’s attachment classifications (p>.04). However, we found significant differences in the scales of the FFI (Table1), as the BE group had higher scores of preoccupation (p=.02) and lower scores of security (p=.05) in attachment representations, showing also poorest narrative coherence (p=.01). Discussion. The attachment patterns showed by the community girls at-risk for binge-eating were consistent with those found in literature on bulimia, more than in anorexia. Authors discussed clinical implications and future directions.