The purpose of this project was to summarize psychotherapy outcomes for eating disorders using meta-analysis. Psychotherapy was defined as any psychosocial treatment including face-to-face therapy, self-help, and Internet approaches. All primary studies, meeting inclusion/exclusion criteria from 1980 to 2010, were included. Results suggested that individuals treated with active treatments demonstrate better outcomes than those in control conditions (d= 0.33, p <.01, 95% CI [0.19-0.46]). CBT was the most often occurring treatment in the primary studies and a small effect, favoring CBT, was found when the treatment was compared to all other active treatments (d = 0.16, p = .02, 95% CI [0.03-0.28]). Internet and self-help approaches continue to show promise with Internet treatments demonstrating superior outcomes to control conditions (d = 0.54, p <.01, 95% CI [0.19-0.90]). More research is needed to determine whether these approaches can suffice as stand-alone treatments or if they are best used in addition to already well-established approaches (i.e., individual CBT). The meta-analysis also explored whether treatment type, outcome measure, diagnosis, attrition, and diagnostic severity moderate treatment effect. Finally, the meta-analysis updated and improved upon previous meta-analyses by focusing on randomized controlled trials, including all diagnoses of eating disorder, only combining effect sizes from similar conditions, including all possible studies that meet inclusion criteria, computing and comparing effect sizes for outcome measures beyond those considered primary to eating disorder treatment, and also addressing outcomes for all brands of psychotherapy including traditional talk therapy, self-help, and Internet approaches.



College and Department

Family, Home, and Social Sciences; Psychology



Date Submitted


Document Type





eating disorders, psychotherapy outcome, meta-analysis

Included in

Psychology Commons