Abstract
BACKGROUND: Anorexia nervosa is complex and difficult to treat. In cognitive therapies the focus has been on cognitive content rather than process. Process-oriented therapies may modify the higher level cognitive processes of metacognition, reported as dysfunctional in adult anorexia nervosa. Their association with clinical features of anorexia nervosa, however, is unclear. With reclassification of anorexia nervosa by DSM-5 into typical and atypical groups, comparability of metacognition and drive for thinness across groups and relationships within groups is also unclear. Main objectives were to determine whether metacognitive factors differ across typical and atypical anorexia nervosa and a non-clinical community sample, and to explore a process model by determining whether drive for thinness is concurrently predicted by metacognitive factors. METHODS: Women receiving treatment for anorexia nervosa (n = 119) and non-clinical community participants (n = 100), aged between 18 and 46 years, completed the Eating Disorders Inventory (3(rd) Edition) and Metacognitions Questionnaire (Brief Version). Body Mass Index (BMI) of 18.5 kg/m(2) differentiated between typical (n = 75) and atypical (n = 44) anorexia nervosa. Multivariate analyses of variance and regression analyses were conducted. RESULTS: Metacognitive profiles were similar in both typical and atypical anorexia nervosa and confirmed as more dysfunctional than in the non-clinical group. Drive for thinness was concurrently predicted in the typical patients by the metacognitive factors, positive beliefs about worry, and need to control thoughts; in the atypical patients by negative beliefs about worry and, inversely, by cognitive self-consciousness, and in the non-clinical group by cognitive self-consciousness. CONCLUSIONS: Despite having a healthier weight, the atypical group was as severely affected by dysfunctional metacognitions and drive for thinness as the typical group. Because metacognition concurrently predicted drive for thinness in both groups, a role for process-oriented therapy in adults is suggested. Implications are discussed.