ABSTRACT
Objective
Executive function deficits are well documented in anorexia nervosa and relate to poorer treatment outcomes, but their clinical relevance in bulimia nervosa (BN) is less clear. This study investigated associations between neurocognition and treatment outcomes in adolescents with BN using complementary hypothesis-driven and data-driven approaches.
Methods
Adolescents with BN (N = 82) enrolled in a randomised controlled treatment trial completed the Delis–Kaplan Executive Functioning System (D-KEFS) and the Eating Disorder Examination (EDE). A priori executive functions investigated included cognitive flexibility, inhibition, problem-solving, and planning. K-means clustering derived neurocognitive profiles based on D-KEFS performance. Multilevel models examined whether a priori executive functions and cluster membership predicted trajectories of EDE Global Scores, objective binge-eating, and self-induced vomiting episodes from baseline to end-of-treatment and 6- and 12-month follow-ups, controlling for treatment condition (i.e., family-based treatment [FBT] versus cognitive behavioural therapy [CBT]).
Results
None of the a priori executive function indices predicted any outcome (all ps > 0.050). K-means clustering identified two distinct neurocognitive profiles. Relative to Cluster 1 (high-average D-KEFS performance), individuals in Cluster 2 (low-average D-KEFS performance) exhibited a significantly higher frequency of self-induced vomiting across post-treatment timepoints (p = 0.049).
Discussion
Findings suggest that neurocognitive profiles may be more informative for predicting psychopathology in adolescent BN than individual executive function measures. Better overall performance on the D-KEFS may confer protection against higher purging frequency across time. Identifying neurocognitive profiles at intake may help identify adolescents at risk for elevated purging over time and inform tailored interventions to support treatment outcome.