Prevention of eating disorders: an umbrella review
Systematic review
|Published
Our aim was to conduct an umbrella review (systematic overview of systematic reviews) on the effect of preventive interventions for eating disorders across all age groups.
Download
Key message
Eating disorders are a condition of increasing prevalence that causes considerable suffering for those affected and their families, and prevention therefore has great potential. We conducted an umbrella review (systematic overview of systematic reviews) of interventions for the prevention of eating disorders, based on an evidence and gap map on prevention and early intervention for eating disorders published in 2025 by NIPH.
The methodological quality of the included reviews was assessed using AMSTAR-2 and we based our conclusions on the most recent reviews of high or moderate methodological quality. We evaluated the certainty of the evidence using the GRADE approach (Grading of Recommendations, Assessment, Development and Evaluation).
We identified and assessed more than 20 interventions across universal, selective and indicated prevention. Only one outcome was assessed as having moderate certainty in the results: improvement in body image immediately following the intervention. The intervention was digital body image programmes delivered as a universal preventive intervention to women aged over 17 years. For all remaining outcomes, we have low or very low certainty of the evidence, indicating that we are uncertain about the effect.
The main conclusion is that there is a lack of strong evidence due to small and few well-conducted controlled trials. The large number of different interventions means that the evidence base for each individual intervention is often limited, and research examining prevention of an eating disorder diagnosis is particularly scarce. To draw conclusions, further well-conducted studies examining the same intervention and comparator are needed.
Summary
Introduction
Eating disorders encompass a group of mental disorders characterised by symptoms related to thoughts and behaviours concerning food, body and weight. The prevalence is increasing, and the condition can cause considerable suffering for those affected and their families, as well as substantial socioeconomic consequences. Prevention therefore has great potential. Preventive interventions can be classified as universal (targeting the general population), selective (targeting groups at increased risk) or indicated (targeting individuals showing early signs of eating disorders).
Objective
Our aim was to conduct an umbrella review (systematic overview of systematic reviews) on the effect of preventive interventions for eating disorders across all age groups. The umbrella review was commissioned by the Norwegian Directorate of Health, which is preparing a knowledge summary for the Ministry of Health and Care Services as part of the mandate "Prevention and treatment of eating disorders" (TB2023-36), anchored in the National Action Plan for Mental Health (2023–2033).
Method
This umbrella review builds on an evidence and gap map on prevention and early intervention for eating disorders published in 2025 by the Norwegian Institute of Public Health (NIPH). The literature search and selection of systematic reviews were conducted as part of the evidence and gap map, and encompassed searches in eight databases. Of 2559 identified publications, 78 systematic reviews were included after screening, of which 41 with a search date from 2020 to February 2025 were assessed further. To be included in the umbrella review, the reviews had to be of moderate or high methodological quality, assessed using the AMSTAR-2 checklist as part of the evidence and gap map. To be more inclusive on sex and age groups, three systematic reviews not previously assessed in the evidence and gap map were quality appraised in this umbrella review. Nine systematic reviews met the requirement of moderate or high methodological quality and constitute the evidence base for the results. One team member extracted results, and another verified the extraction. We assessed the certainty of evidence using the GRADE approach (Grading of Recommendations, Assessment, Development and Evaluation) for the outcomes eating disorder diagnosis, body image, body dissatisfaction and disordered eating.
Results
We examined over 20 interventions across universal, selective and indicated prevention. We assessed the certainty in most outcomes as very low, primarily due to small and few well conducted primary studies, considerable heterogeneity in intervention content and delivery, and sparse reporting of effect estimates for several outcomes.
We have moderate certainty in one outcome for universal preventive interventions:
- Digital body image programmes probably lead to improvement in body image compared with control or no intervention, measured immediately after the intervention was delivered to women aged over 17 years.
We have low certainty in several outcomes for universal preventive interventions:
- Unguided online psychoeducation (eBodyProject) may have a small positive effect on body image, body dissatisfaction and disordered eating compared with brochure or video, immediately after the intervention and at short- and long-term follow-up for women with a mean age of approximately 20 years.
- Cognitive behavioural therapy may have little or no effect on body dissatisfaction compared with usual education or waitlist for boys under 18 years.
- Cognitive dissonance-based therapy may lead to improvement in body image and body dissatisfaction compared with waitlist, immediately after the intervention and at 1 to 6 months follow-up for men over 18 years.
For selective and indicated preventive interventions, yoga was the only intervention examined in these settings alone. We have very low certainty in these results.
We have low certainty in several outcomes for universal, selective and indicated preventive interventions examined jointly:
- Body image flexibility interventions (BIF) may lead to lower body dissatisfaction compared with no or minimal intervention, at follow-up from 1 week to 6 months and after exposure to a body image threat immediately after the intervention for adolescents and young adults aged 12 to 25 years (predominantly women)
- Body image flexibility interventions (BIF) may lead to improved body image and possibly little or no change in body dissatisfaction compared with evidence-based interventions (EBI), measured immediately after the intervention for adolescents and young adults aged 12 to 25 years (predominantly women)
We have low certainty in several outcomes for universal and indicated preventive interventions examined jointly:
- Digital self-help interventions may lead to little or no change in disordered eating compared with waitlist or video, both immediately after the intervention and at follow-up for adolescents and young adults aged 13 to 24 years (predominantly women)
We have low certainty in several outcomes for selective and indicated preventive interventions examined jointly:
- Unguided digital self-help platforms focusing on cognitive behavioural therapy may lead to improvement in disordered eating compared with no intervention, immediately after the intervention for young adults (predominantly women) at risk of developing or with established symptoms of an eating disorder
- Unguided digital self-help platforms focusing on dissonance-based therapy may lead to improvement in disordered eating compared with no intervention, immediately after the intervention for women with a mean age of 20 years at risk of developing or with established symptoms of an eating disorder
For the remaining interventions, we have very low certainty in the results. Very little research has examined prevention of a clinical eating disorder diagnosis as an outcome. Only two reviews examined this, both concerning universal psychoeducation. We have very low certainty in the results and are therefore uncertain about the effect of psychoeducation on the risk of developing a clinical diagnosis. For the remaining interventions, the effect on eating disorder diagnosis has not been examined.
Discussion
We have systematically summarised nine systematic reviews of moderate methodological quality examining the effectiveness of preventive interventions for eating disorders. Overall, this umbrella review shows that the evidence base is fragmented and generally weak. Most interventions have very low or low certainty of evidence, and the review highlights substantial knowledge gaps. Several systematic reviews combine universal, selective and indicated interventions, making it difficult to assess which interventions work for which target groups and at which stages of the risk trajectory. There is limited research on the diagnosis outcome, long-term effects and minority groups, and relatively few interventions have been examined in female populations despite the higher prevalence of eating disorders among women.
Methodological weaknesses and varying quality in the reporting of results contribute to further uncertainty regarding the effectiveness of these interventions. Taken together, the results indicate that newer, larger and well-designed studies are needed before clear recommendations for preventive practice can be made. This umbrella review therefore primarily provides a basis for further research, rather than definitive conclusions about effective interventions.
Conclusion
Our main conclusion is that there is a lack of solid documentation on the effectiveness of preventive interventions for eating disorders.