Eating disorder treatment programs compared to controls for adolescents and young adults: Systematic review and three-level meta-analyses
- Forfatter(e)
- Arbit, S. E., Kirsch, A. C., Nowakowski, L., Meagher, M. F., Karl, A. N., Blum, A. L., Assink, M., Conley, C. S.
- År
- 2026
- Tidsskrift
- Eating Behaviors
- Volum
- 60
- Sider
- 102070
- Kategori(er)
- Spiseforstyrrelser
- Tiltakstype(r)
- Kognitiv atferdsterapi, atferdsterapi og kognitiv terapi
- Abstract
Rates of eating disorders are on the rise among young people, and there is growing concern about the ability of existing treatment options to meet the evolving needs of this population. This underscores the urgent need to evaluate (a) the current state of the literature, and (b) the effectiveness of interventions tailored to adolescents and young adults (AYAs), a population uniquely vulnerable to the onset and chronic progression of untreated eating disorders. This paper presents a systematic and meta-analytic review of psychological (non-biological/pharmacological) interventions compared to control conditions for eating disorders (EDs) in AYAs. A systematic search was conducted in PsycInfo, PubMed, and ProQuest Dissertations and Theses for studies published between January 1980 and March 2024. Supplemented by various other search methods, we identified 53 interventions, contained within 41 reports, yielding an overall sample size of 3140. Studies were evaluated using elements of the Cochrane risk of bias tool. Analyses were conducted using three-level, random effects models. Results indicated that, compared to controls, interventions targeting anorexia nervosa (AN), bulimia nervosa (BN), binge eating disorder (BED), and a blend of ED diagnoses among AYAs were effective for a range of ED outcomes, with maintenance of effects at follow-up: Hedges' g across interventions targeting AN = 0.202 (0.051; 0.352), BN = 0.467 (0.241; 0.693), BED = 0.601 (0.294; 0.908), and a blend of diagnoses = 0.378 (0.111; 0.644). Programs targeting BN and multiple diagnoses were similarly effective for non-ED outcomes. Differential effects emerged for the type of control condition such that interventions were most effective when compared to no intervention or waitlist controls. There was no moderating effect for type of outcome, duration of diagnosis, age, use of technology, duration of intervention, duration of follow-up, publication year, or risk of bias. Evidence is limited by the small number of studies and sample sizes within each ED diagnosis, and points to the need for development and refinement of existing interventions. Clinical implications and recommendations for future research on eating disorder interventions for AYAs are highlighted. This study was registered with Open Science Foundation (osf.io/wc6m7) and funded by internal research grants from Loyola University Chicago.