Feasibility, Efficacy, and Predictors for Treatment Response in a Tailored Digital CBT-E Program for Binge Eating in Adolescents and Young Adults: A Pilot Randomized Controlled Trial With Multimodal Assessment (Preprint)

BACKGROUND Binge eating is a prevalent and clinically significant condition in adolescents and young adults, associated with functional impairment and increased risk of progression to full-syndrome eating disorders. Digital guided self-help programs based on enhanced cognitive behavioral therapy (CBT-E) have been proposed as scalable options for early intervention; however, implementing CBT-E’s individualized formulation in digital formats remains challenging. In addition, objective indicators of self-regulatory change underlying CBT-E effects are limited. OBJECTIVE This pilot randomized controlled trial evaluated the feasibility and preliminary outcomes of PlanEat, a developmentally and symptomatically tailored digital CBT-E–based guided self-help program. Using a multimodal assessment approach, we examined neurophysiological indicators of self-regulation and predictors of treatment response. METHODS Adolescents and young adults aged 13–22 years with recent binge-eating episodes were randomized to an immediate intervention group or a waitlist control group in a 2-arm waitlist-controlled crossover design. The intervention consisted of an 8-week mobile-based guided self-help CBT-E program with symptom-driven tailoring. Feasibility outcomes included attrition, adherence, engagement, and usability. Clinical outcomes were assessed using validated self-report measures of eating disorder psychopathology and functional impairment. Neurophysiological outcomes included functional near-infrared spectroscopy (fNIRS) during executive control tasks and heart rate variability (HRV) during a virtual reality–based social stress paradigm. Intervention effects during the active treatment period (T0–T1) were analyzed using linear mixed-effects models and covariance analysis. Hierarchical regression analyses explored whether adherence and baseline neurophysiological indicators predicted change in global Eating Disorder Examination Questionnaire (EDE-Q) scores. RESULTS Of 64 randomized participants, 46 were included in the primary clinical analyses. Feasibility criteria were met, with favorable overall usability ratings. The immediate intervention group showed a significantly greater reduction in global EDE-Q scores compared with the waitlist group (β = 0.66, SE = 0.28, P = .021). Improvements were also observed across eating-related outcomes, functional impairment, and self-esteem. Clinical benefits were largely maintained at 16-week follow-up, with additional reductions in anxiety symptoms. Exploratory neurophysiological analyses showed attenuated reductions in prefrontal activation during inhibitory control tasks in the intervention group and a significant Group × Time interaction in lnHF reactivity, indicating divergent autonomic stress regulation trajectories. In hierarchical regression models, greater adherence and lower baseline upper right prefrontal HbR activation were independently associated with larger reductions in global EDE-Q scores. CONCLUSIONS The PlanEat program showed acceptable feasibility and preliminary clinical benefits in adolescents and young adults with binge eating. Exploratory multimodal findings suggest that neurophysiological measures may complement behavioral indicators by providing objective indices of self-regulatory change and prognostic information. Together, these results support further evaluation of PlanEat in a fully powered randomized controlled trial and highlight the potential value of integrating multimodal markers to better characterize treatment response in digital CBT-E interventions. CLINICALTRIAL ClinicalTrials.gov Identifier: NCT05615090