Remotely Delivered Cognitive Behavioral Therapy for Adults With an Eating Disorder: Retrospective Analysis of a Real-World Patient Sample

Abstract

Background: Enhanced cognitive behavioral therapy (CBT-E) is the prevailing treatment approach for adult eating disorders. CBT-E is a variant of cognitive behavioral therapy, modified specifically to treat an eating disorder. Systematic reviews have established the effectiveness of CBT-E for adults when delivered face to face. However, few studies have evaluated evidence-based eating disorder treatment outcomes for programs intentionally designed to be delivered remotely. Objective: The objective of this study was to examine the clinical utility of CBT-E for adults with eating disorders using data from a national treatment program designed specifically for remote delivery. Methods: This was a pre-post observational cohort study conducted in a naturalistic setting where patients received treatment through standard clinical pathways, including typical referral, intake, and treatment processes. The participant sample for the study was identified through retrospective chart review and included adult patients (aged ≥18 y) diagnosed with anorexia nervosa, bulimia nervosa, binge eating disorder, or other specified feeding or eating disorder. For adult patients with these diagnoses, CBT-E is generally the first line of care in the program. CBT-E was developed to be transdiagnostic, and rather than focusing on a specific diagnosis, treatment focuses on treating the problematic beliefs related to weight, shape, and eating that maintain the eating disorder. CBT-E is highly individualized, and the treatment provider creates a treatment plan to match the specific eating disorder symptoms experienced by the patient. The recommended cadence of sessions is weekly. The criterion of utility was the magnitude and consistency of symptom change in weight gain and eating disorder, depression, and anxiety symptoms during CBT-E treatment. Survival analyses assessed patient and treatment characteristics. Multilevel models assessed the changes in outcomes both over time and at weeks 20 and 40, as these time points generally aligned with CBT-E clinical trial end points. Results: The patient sample (N=1629) predominantly consisted of White (n=1166, 71.6%), cisgender women (n=1403, 86.1%), with a mean age of 30 (SD 12) years. The overall median length of stay was 22 (95% CI 20-25) weeks. In all, 416 (25.5%) patients required weight restoration. The estimated probability of achieving weight restoration was 0.50 (95% CI 0.43-0.57) just before week 40 of treatment. By 40 weeks of treatment, the probability of achieving subclinical status for eating disorder symptoms was 0.48 (95% CI 0.44-0.51); for depression, the probability was 0.55 (95% CI 0.51-0.59), and for anxiety, the probability was 0.56 (95% CI 0.51-0.60). Time in treatment was significantly associated with improved symptoms across all outcomes (all P