Predictors and moderators of three online interventions for eating disorder symptoms in a randomized controlled trial

Pieter J. Rohrbach*, Marjolein Fokkema, Philip Spinhoven, Eric F. Van Furth, Alexandra E. Dingemans

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


To optimize treatment recommendations for eating disorders, it is important to investigate whether some individuals may benefit more (or less) from certain treatments. The current study explored predictors and moderators of an automated online self-help intervention “Featback” and online support from a recovered expert patient.

Data were used from a randomized controlled trial. For a period of 8 weeks, participants aged 16 or older with at least mild eating disorder symptoms were randomized to four conditions: (1) Featback, (2) chat or e-mail support from an expert patient, (3) Featback with expert-patient support, and (4) a waitlist. A mixed-effects partitioning method was used to see if age, educational level, BMI, motivation to change, treatment history, duration of eating disorder, number of binge eating episodes in the past month, eating disorder pathology, self-efficacy, anxiety and depression, social support, or self-esteem predicted or moderated intervention outcomes in terms of eating disorder symptoms (primary outcome), and symptoms of anxiety and depression (secondary outcome).

Higher baseline social support predicted less eating disorder symptoms 8 weeks later, regardless of condition. No variables emerged as moderator for eating disorder symptoms. Participants in the three active conditions who had not received previous eating disorder treatment, experienced larger reductions in anxiety and depression symptoms.

The investigated online low-threshold interventions were especially beneficial for treatment-naïve individuals, but only in terms of secondary outcomes, making them well-suited for early intervention. The study results also highlight the importance of a supportive environment for individuals with eating disorder symptoms.

Public Significance
To optimize treatment recommendations it is important to investigate what works for whom. For an internet-based intervention for eating disorders developed in the Netherlands, individuals who had never received eating disorder treatment seemed to benefit more from the intervention than those who had received eating disorder treatment, because they experienced larger reductions in symptoms of depression and anxiety. Stronger feelings of social support were related to less eating disorder symptoms in the future.
Original languageEnglish
Pages (from-to)1909-1918
JournalInternational Journal of Eating Disorders
Issue number10
Early online date10 Jul 2023
Publication statusPublished - 5 Oct 2023


  • eHealth
  • eating disorders
  • expert patient
  • internet-based intervention
  • moderators
  • predictors
  • randomized controlled trial
  • recursive partitioning


Dive into the research topics of 'Predictors and moderators of three online interventions for eating disorder symptoms in a randomized controlled trial'. Together they form a unique fingerprint.

Cite this