Dietary restraint, binge eating and adherence to an adolescent intensive behavioural weight management intervention: a Fast Track to Health sub-study.

PURPOSE: Dietary restraint often increases or remains unchanged following adolescent weight management interventions. It is unknown whether dietary restraint change in such interventions is associated with binge eating risk or intervention adherence. This study examined associations between (1) dietary restraint, measured using the Eating Disorder Examination Questionnaire (EDE-Q) and Dutch Eating Behaviour Questionnaire (DEBQ); (2) changes in dietary restraint and binge eating; and (3) changes in dietary restraint, anthropometry, and adherence throughout a weight management intervention. METHODS: Secondary data analysis of a clinical trial of behavioural weight management interventions involving adolescents (13-17 years) with obesity and related complications. Associations between dietary restraint, binge eating, and BMI z-score were examined using Pearson or Spearman's rank correlation. Differences in dietary restraint change between adherers and non-adherers were examined by independent samples t-tests or Mann-Whitney U tests. RESULTS: 141 adolescents (14.8 years, 49.6% female) were recruited; 136 had baseline, 130 week-4, 120 week-16, and 92 week-52 data included in analyses. EDE-Q and DEBQ restraint scores were correlated at all timepoints (r = 0.442-0.579, all p < 0.001). Changes in dietary restraint and binge eating from baseline to week-4 and week-16 were not correlated. Week-52 change in DEBQ dietary restraint was associated with change in binge eating (r = 0.347, p < 0.001). Correlations between changes in dietary restraint and BMI z-score and differences in change in dietary restraint between adherers and non-adherers were not significant. CONCLUSION: The findings do not provide robust evidence of dietary restraint as a marker of adherence or binge eating risk during a behavioural weight management intervention. LEVEL OF EVIDENCE: Level III: Evidence obtained from well-designed cohort or case-control analytic studies.