Abstract
Background/Aim: Dropouts in intervention trials can affect the statistical power of the study. The aim was to assess predictors of dropping out in a weight loss trial. Methods: Each subject (n = 114) was instructed to follow a diet, energy-restricted by 30% from estimated energy expenditure (approximately 600 kcal/day), for 8 consecutive weeks. A detailed meal plan as well as recipe booklets and instructions were provided by a nutritionist. Results: Antidepressants were used by 19.2% of the dropouts, compared with 1.8% of the completers (p < 0.001). Additionally, those dropping out consumed 35% more soft drinks in total and 44% more diet (artificially sweetened) soft drinks at baseline (habitual intake) than those completing the trial (p = 0.012). Dropouts (18.6%) were on average 1.9 years younger than those completing the study, but the difference was of borderline significance (p = 0.085). Conclusions: The use of antidepressants and heavy habitual soft drink consumption might be considered as exclusion criteria when preparing weight loss intervention trials including dietary restrictions.