OBJECTIVE:To compare the long-term compliance and effects of two low-fat diets differing in carbohydrate to protein ratio on body composition and biomarkers of cardiovascular disease risk in obese subjects with hyperinsulinemia. DESIGN: Outpatient, parallel, clinical intervention study of two groups of subjects randomly assigned to either a standard protein (SP; 15% protein, 55% carbohydrate) or high-protein (HP; 30% protein, 40% carbohydrate) diet, during 12 weeks of energy restriction (B6.5 MJ/day) and 4 weeks of energy balance (B8.3 MJ/day). Subsequently, subjects were asked to maintain the same dietary pattern for the succeeding 52 weeks with minimal professional support. SUBJECTS: A total of 58 obese, nondietetic subjects with hyperinsulinemia (13 males/45 females, mean age 50.2 y, mean body mass index (BMI) 34.0 kg/m 2 , mean fasting insulin 17.8 mU/l) participated in the study. MEASUREMENTS: Body composition, blood pressure, blood lipids, fasting glucose, insulin, CRP and sICAM-1 were measured at baseline and at weeks 16 and 68. Urinary urea/creatinine ratio was measured at baseline, week 16 and at 3 monthly intervals thereafter. RESULTS: In total, 43 subjects completed the study with similar dropouts in each group (P ¼ 0.76). At week 68, there was net weight loss (SP À2.973.6%, HP À4.175.8%; Po0.01) due entirely to fat loss (Po0.001) with no diet effect. Both diets significantly increased HDL cholesterol concentrations (Po0.001) and decreased fasting insulin, insulin resistance, sICAM-1 and CRP levels (Po0.05). Protein intake was significantly greater in HP during the initial 16 weeks (Po0.001), but decreased in HP and increased in SP during 52-week follow-up, with no difference between groups at week 68, indicating poor long-term dietary adherence behaviour to both dietary patterns. CONCLUSION: Without active ongoing dietary advice, adherence to dietary intervention is poor. Nonetheless, both dietary patterns achieved net weight loss and improvements in cardiovascular risk factors.