Measuring and modelling maladaptive core beliefs in eating disorders

Abstract

Maladaptive core beliefs have been implicated in both the development and maintenance of eating disorders (EDs) and have been associated with increased eating psychopathology.

Despite advances in treatment, low remission rates, high comorbidity, and diagnostic migration highlight the need to examine higher-order cognitive factors that contribute to EDs. This thesis explored the assessment and role of maladaptive core beliefs in disordered eating, and addressed limitations in measurement tools and theoretical models. Chapter 1 provided a critical narrative review of the literature on measuring and modelling maladaptive core beliefs in EDs and Chapter 2 presented a systematic review of psychometric properties of existing self-report questionnaires assessing ED-related cognitions. Chapter 3 evaluated the 40-item Eating Disorder Core Beliefs Questionnaire (ED-CBQ) and introduced a brief, psychometrically robust version (the 15-item ED-CBQ-R) that assesses four belief dimensions: self-loathing, unassertive/inhibited, demanding, and abandoned/isolated. Chapter 4 validates the ED-CBQ-R in an undergraduate sample, and Chapter 5 established its reliability, construct validity, and clinical utility in treatment-seeking adults across Australia and New Zealand with an ED. This study found that individuals with anorexia nervosa (restricting subtype) reported higher self-loathing than those with binge eating disorder, with self-loathing showing the strongest, consistent associations with symptom severity. Chapter 6 developed a cohesive theoretical model of maladaptive core beliefs in disordered eating, outlining distinct pathways to restriction, binge eating, and compensatory behaviours. Chapter 7 partially validates this model in an international clinical sample. The final chapter synthesised the findings of this thesis, and emphasised the theoretical and clinical significance of core beliefs in EDs, highlighting their potential as key targets for intervention.