4 - (SYM 122) Processes of Change in Residential Transdiagnostic Cognitive-behavioral Therapy for Severe Eating Disorders

Introduction: Eating disorders (EDs) affect millions of Americans annually and are among the most medically compromising psychological disorders (Smink et al., 2013). Residential or inpatient treatment is indicated for individuals with more severe and complex EDs (Twohig et al., 2015). Routine residential contexts increasingly track treatment quality and patient outcomes (Thompson-Brenner et al., 2021), yet research on processes of change is lacking. Within the context of a residential treatment program for severe EDs that has adopted a transdiagnostic CBT model, this exploratory study aims to examine change over time in three theoretically and empirically relevant change variables: (1) emotion avoidance; (2) adaptive coping skill use; and (3) relational connection.

Method: Participants were adolescent or adult female patients (N = 730) presenting for routine ED residential care (Mage = 24.51 years; SD = 9.96). Most patients identified as White (83%); 6% identified as Hispanic. Primary DSM-5 diagnoses included: 44.6% anorexia nervosa; 27.4% bulimia nervosa; 4.1% binge ED; and 17.6% other specified feeding or ED. Participants completed the Progress Monitoring Tool for EDs (PMED; Espel-Huynh et al., 2020), including emotion avoidance, coping skills and relational connection subscale items at program admission, discharge, and six months following discharge (6MFU). We will report descriptive results pertaining to change over time on these process variables, as well as multilevel growth models examining the trajectory of change from admission to 6MFU.

Results: Participants demonstrated statistically significant improvement in emotion avoidance between admission and discharge, d = 1.64 (95% CI = 1.46, 1.82), as well as between admission and 6MFU, d = 1.15 (95% CI = 0.99, 1.31). Participants demonstrated statistically significant improvement in coping skills between admission and discharge, d = 1.94 (95% CI = 1.76, 2.14), yet level of overall improvement in coping skills was not statistically significant between admission and 6MFU, d = 0.11 (95% CI = -0.04, 0.26). A similar pattern was observed for relational connect at discharge (d = 1.51, 95% CI = 1.34, 1.67), and 6MFU (d = 0.01, 95% CI = -0.14, 0.16), indicating a reversal of some within-treatment gains. In addition to these descriptive results, we will present results from a series of longitudinal models exploring change patterns through 6MFU.

Discussion: Results will shed light on processes of change in residential CBT treatment program for severe EDs.