41 Predictors of outcome for cognitive behaviour therapy in binge eating disorder 2 It should, however, be noted that our sample consisted of relatively few low-educated patients, which could have limited the predictive power for education. Apart from this, our analyses revealed several other interesting findings. First of all, a higher level of drive for thinness at baseline predicted better outcome in the short term and remained a post-treatment predictor for follow-up. This robust finding seems remarkable, as a high drive for thinness among women with lifetime diagnosis of bulimia nervosa (BN) has been found to be associated with a decreased likelihood of recovery (Keski-Rahkonen et al., 2012). When looked at more closely, however, the drive for thinness subscale of the EDI is likely to measure something different in BN patients with an average weight within the normal range than in obese BED patients. Firstly, drive for thinness scores are lower in obese BED patients than in BN patients (Brewin et al., 2014; Jordan et al., 2014). Secondly, obese people with and without BED do not differ on drive for thinness (Ramacciotti et al., 2008). Some preoccupation with thinness and with fear of becoming fat in obese BED patients seems to reflect a realistic way of interpreting the risks associated with obesity and can therefore be an extra motivation to stop binge eating. In addition, higher pretreatment scores on general psychopathology (as measured by the SCL-90 total score) predict more post-treatment binge eating pathology. It is likely that higher general levels of psychopathology at the start interfere with addressing binge eating in treatment. At the same time however, higher post-treatment scores on general psychopathology predict less binge eating pathology at follow-up in women. We do not currently understand how this relationship can be explained. Future studies should show whether this is a robust finding. Furthermore, the level of depression (as measured with the BDI) did not predict binge eating pathology even though there was a wide range of depression severity within our sample. At the same time, however, more depressive complaints at baseline (as measured by the subscale depression of the SCL-90, which is highly correlated with the BDI in our sample) did predict lower levels of binge eating pathology at the end of treatment. Our results therefore underline the ambiguous predictive relationship between levels of depression and treatment outcome in BED (Grilo et al, 2012; Masheb & Grilo 2008; Ricca et al., 2010; Wilson et al, 2010). Additionally, higher post-treatment scores on insufficiency were found to predict higher follow-up binge eating pathology. Insufficiency, as measured by the SCL-90, relates to thoughts, impulses and problems in the execution of behaviour and focuses on the cognitive domain (Arrindell & Ettema, 2003). There are some, albeit inconclusive, indications that both people with bulimic eating disorders and people with obesity show
RkJQdWJsaXNoZXIy MjY0ODMw