Objective: To investigate whether a hypo-energetic low-fat diet is superior to a hypo-energetic high-fat diet for the treatment of obesity. Design: Open-label, 10-week dietary intervention comparing two hypo-energetic (À600 kcal/day) diets with a fat energy percent of 20-25 or 40-45. Subjects: Obese (BMI X30 kg/m 2 ) adult subjects (n ¼ 771), from eight European centers. Measurements: Body weight loss, dropout rates, proportion of subjects who lost more than 10% of initial body weight, blood lipid profile, insulin and glucose. Results: The dietary fat energy percent was 25% in the low-fat group and 40% in the high-fat group (mean difference: 16 (95% confidence interval (CI) 15-17)%). Average weight loss was 6.9 kg in the low-fat group and 6.6 kg in the high-fat group (mean difference: 0.3 (95% CI À0.2 to 0.8) kg). Dropout was 13.6% (n ¼ 53) in the low-fat group and 18.3% (n ¼ 70) in the high-fat group (P ¼ 0.001). Among completers, more subjects lost 410% in the low-fat group than in the high-fat group ((20.8%, n ¼ 70) versus (14.7%, n ¼ 46), P ¼ 0.02). Fasting plasma total, low-density lipoprotein-and high-density lipoprotein-cholesterol decreased in both groups, but more so in the low-fat group than in the high-fat group. Fasting plasma insulin and glucose were lowered equally by both diets. Conclusions: The low-fat diet produced similar mean weight loss as the high-fat diet, but resulted in more subjects losing 410% of initial body weight and fewer dropouts. Both diets produced favorable changes in fasting blood lipids, insulin and glucose.