Abstract
Eating disorders (EDs) are undertreated worldwide. In the UK the lag between recognition of symptoms and treatment ranges from about 15 months to in excess of 2 years. Internet-based cognitive behaviour therapy (ICBT) could be a viable alternative to face-to-face cognitive behaviour therapy (CBT) that avoids the negative impacts of delayed interventions. Based on evidence from randomised controlled trials (RCTs), this systematic review investigated the efficacy of minimally guided self-help ICBT, without face-to-face therapy, for the prevention, treatment and relapse prevention of all types of EDs in adults. The electronic databases MEDLINE, PsychINFO, CENTRAL, Scopus, and Web of Science were searched between 1991 and 2021. Inclusion criteria specified RCTs with ICBT versus inactive comparison groups. The Cochrane Risk of Bias Tool-2 was used for quality assessments. Qualitative synthesis and meta-analyses were conducted. Findings typically showed medium significant beneficial effect sizes for prevention studies ranging from (-0.31 [95% CI: -0.57, -0.06] to -0.47 [95% CI: -0.82, -0.11]) and generally large effect sizes for the treatment studies ranging from (-0.30 [95% CI: -0.57, -0.03] to -1.11 [95% CI: -1.47, -0.75]). Relapse prevention studies yielded mainly small non-significant beneficial effects with significant effect sizes of (-0.29 [95% CI: -0.56, -0.03] and -0.43 [95% CI: -0.70, -0.16]). Only the treatment studies reached clinical significance and cognitive symptoms improved more than behavioural symptoms. ICBT appears to be efficacious for the prevention, treatment and relapse prevention of eating disorders with treatment interventions being the most beneficial. However, the evidence base is very small, particularly for treatment and relapse prevention, indicating the need for more high-quality RCTs.