Are there differences in bingeing/purging and restrictive subtypes of anorexia nervosa on central coherence and set shifting?

Background: Anorexia nervosa (AN) has been associated with dysfunctions in central coherence (CC) and set shifting (SS). These associations have been repeatedly confirmed for the total group of AN, while eventual differences between the subtypes of AN (restrictive and bingeing/purging types) have not been profoundly investigated up to now. The aim of this study is to further examine this topic. We hypothesize that the restrictive AN patients will exhibit lower levels of performance than the bingeing/purging AN patients. In addition, we sought to examine the relationship between performance on CC/SS measures and clinical/demographical features. Methods: Sample: This study contains 83 participants: 31 with AN restrictive type (AN-R), 21 with AN bingeing/purging type (AN-BP) and 31 healthy controls (HC). Measures: All participants completed a neuropsychological test battery individually, including two measures of CC (Picture Completion and Block Design) and three measures of SS (a designed task-switching paradigm, the Wisconsin Card Sorting Test and the Trail Making Test). Clinical characteristics were assessed using self report questionnaires, including the Beck Depression Inventory, the State-Trait Anxiety Inventory, and the Frost Multidimensional Perfectionism Scale. Interviewing was used to gather demographical characteristics. Statistics: Groups were compared on neuropsychological functioning using one-way ANOVA’s with Tukey correction for multiple comparison. The Kruskal-Wallis statistic was employed when data were not normally distributed. Pearson’s correlations were used to investigate links with clinical and demographical variables. Results: Significant differences between the subtypes of AN were found on the two measures of CC, more precisely on Block Design (F(2)=16,14; p=.008) and Object Assembly (F(2)=5,79; p=.008). A worse performance was found for the AN-R patients compared to the results of the AN-BP patients, while these had comparable results with the HC subjects (with p>.05). These group differences persist, adding age as a covariate (F(1)=5,88; p=.019). On SS measures, no group differences could be demonstrated among clinical groups or with HC subjects. Age, current BMI, lowest BMI after the age of 16, education (in years) and level of depressive and anxiety symptoms were not correlated with performance on CC and SS measures (p>0.5). Conclusions: AN-R patients seem to differ from AN-BP patients on CC, with AN-R performing worse, while the AN-BP patients did not show a CC deficit. No deficit on SS was found for any AN subtype. These results support the idea that the two subtypes of AN need a different approach in cognitive remediation.