Quality of randomized controlled trials in eating disorder prevention.

OBJECTIVE: To investigate the quality of randomized controlled trials (RCTs) of eating disorder prevention. METHOD: A systematic literature search was conducted in Medline, PsycINFO, Embase, Scopus, and the Cochrane Collaboration Library to January 2016. Studies were included if they were RCTs that tested an eating disorder prevention program. We identified 96 studies with a total 15,350 participants (91% female, M age = 17 years) and rated quality with the Quality Rating Scale (QRS; Moncrieff et al., 2001). RESULTS: The mean QRS score was 62% (SD = 13%). Several standards of quality were not frequently fulfilled (i.e., failed to achieve an optimal rating), for example, power calculation (85%), intent-to-treat analysis (54%), blinding of assessor (75%), representative sample (78%), adequate sample size (75%), and appropriate duration of trial including follow-up (67%). QRS was positively and significantly associated with publication year, number of authors, and PubMed-indexation. DISCUSSION: Given the majority of eating disorder prevention studies had problems with trial quality, it is recommended that future RCTs follow quality checklists and CONSORT guidelines, that RCTs are registered, and protocols published in advance. In addition, funding bodies are called on to deliver the support needed to ensure that preventions for eating disorders are efficiently and cost-effectively achieved.

Prevention of eating disorders: A systematic review of randomized, controlled trials.

OBJECTIVE: This systematic review evaluated the efficacy of universal, selective, and indicated eating disorder prevention. METHOD: A systematic literature search was conducted in Medline, PsycINFO, Embase, Scopus, and the Cochrane Collaboration Library databases to January 2016. Studies were included if they were randomized, controlled trials (RCT) and tested an eating disorder prevention program. We retrieved 13 RCTs of universal prevention (N = 3,989 participants, 55% female, M age = 13.0 years), 85 RCTs of selective prevention (N = 11,949 participants, 99% female, M age = 17.6 years), and 8 RCTs of indicated prevention (N = 510 participants, 100% female, M age = 20.1 years). Meta-analysis was performed with selective prevention trials. As there were a limited number of universal and indicated trials, narrative synthesis was conducted. RESULTS: Media literacy had the most support for universal prevention. Most universal approaches showed significant modest effects on risk factors. Dissonance-based was the best supported approach for selective prevention. Cognitive-behavior therapy (CBT), a healthy weight program, media literacy, and psychoeducation, were also effective for selective prevention and effects were maintained at follow-up. CBT was supported for indicated prevention and effects were maintained at follow-up. DISCUSSION: The modest effects for universal prevention were likely due to floor effects. The evidence for selective prevention suggests that empirically supported approaches should be disseminated on a wider basis. Our findings suggest CBT should be offered for indicated populations. Overall, results suggest efficacy of several prevention programs for reducing risk for eating disorders, and that wider dissemination is required. © 2016 Wiley Periodicals, Inc.