Predictors, moderators, and mediators of treatment outcome following manualised cognitive-behavioural therapy for eating disorders: A systematic review

This systematic review synthesised the literature on predictors, moderators, and mediators of outcome following Fairburn's CBT for eating disorders. Sixty-five articles were included. The relationship between individual variables and outcome was synthesised separately across diagnoses and treatment format. Early change was found to be a consistent mediator of better outcomes across all eating disorders. Moderators were mostly tested in binge eating disorder, and most moderators did not affect cognitive-behavioural treatment outcome relative to other treatments. No consistent predictors emerged. Findings suggest that it is unclear how and for whom this treatment works. More research testing mediators and moderators is needed, and variables selected for analyses need to be empirically and theoretically driven. Future recommendations include the need for authors to (i) interpret the clinical and statistical significance of findings; (ii) use a consistent definition of outcome so that studies can be directly compared; and (iii) report null and statistically significant findings.