4 - (SYM 1) Reward Retraining: A Pilot Randomized Controlled Trial of a Novel Treatment Approach for Transdiagnostic Binge Eating

Abstract Body

Background: Low reward response to conventionally rewarding stimuli and high reward response to food (i.e., reward imbalance), has been supported as a maintenance factor for binge-spectrum eating disorders. The current study was a pilot RCT testing a novel treatment approach for binge eating targeting reward imbalance, called Reward Re-Training (RRT), in comparison to a therapeutic attention control condition (supportive psychotherapy; ST). We hypothesized that RRT would have high feasibility and acceptability (Primary Aim 1), produce significantly greater improvement in palatable food reward, day-to-day activity reward, and social support compared to supportive therapy which would mediate differences in primary outcomes (Primary Aim 2), and result in significant reduction of eating disorder symptoms at posttreatment and 3-month follow-up (Primary Aim 3).

Methods: Fifty-nine participants were randomly assigned to receive either 10 group sessions of RRT or ST via videoconferencing software. Assessments of eating pathology and treatment targets were completed at pretreatment, mid-treatment, and posttreatment, and 3-month following the end of treatment. Multilevel models (MLMs) were conducted for Aim 2 and Aim 3.

Results: Feasibility and acceptability success benchmarks were achieved. Results found a significant indirect effect of RRT on lower posttreatment global eating pathology through decreases in reward to food from pretreatment to mid-treatment. No significant difference was observed between groups in terms of change in binge frequency or global eating pathology. In terms of effect sizes, MLMs found nonsignificant and small effects of group on change in binge eating frequency (F(4) = −0.65, p = .58, f2 = 0.10). Descriptive statistics indicate that, while not statistically significant, RRT had slightly lower binge eating frequency at mid-treatment compared to ST. At posttreatment and follow-up, both groups demonstrated similar binge eating frequency. MLMs also found nonsignificant and small effects of group on change in global eating pathology scores (F(4) = 2.02, p = .14, f2 = 0.12). RRT and ST reported similar global eating pathology scores at baseline and posttreatment, but RRT reported slightly lower global eating pathology scores at follow-up than ST.

Conclusion: The current study supports the feasibility and acceptability of RRT, and the preliminary efficacy of both RRT and ST. Further research comparing these approaches to CBT would help to inform who might benefit from alternative treatment approaches.