The aim of the study was to establish the best cut-off value for the homeostatic model assessment (HOMA) index in identifying children and adolescents with the metabolic syndrome. The study included 72 non-obese and 68 obese children aged 7 to 16 years. Obesity is defined using the criteria proposed by Cole et al., being included as metabolic syndrome variables waist circumference, systolic blood pressure, diastolic blood pressure and seric values of glucose, uric acid, fasting insulin, leptin, triglycerides and HDL-cholesterol. Children were considered as having the metabolic syndrome when four or more characteristics showed abnormal values. The HOMA index was calculated as the product of the fasting plasma insulin level (microU/mL) and the fasting plasma glucose level (mmol/L), divided by 22.5. HOMA index cut-offs from the 5th to the 95th percentile were used. A receiver operating characteristic (ROC) curve was generated using the different HOMA cut-offs for the screening of the metabolic syndrome. The areas under the ROC curve, 95% confidence intervals, and the point to the ROC curve closest to 1, were calculated. The area under the ROC curve was 0.863 (95% C.I.: 0.797, 0.930). The point closest to 1 corresponds to the 60th percentile of the HOMA index distribution in our sample. HOMA index value at the 60th percentile was 2.28. Cut-off values corresponding to a range of HOMA index from the 50 to the 75 percentile, showed similar distances to 1. HOMA index values for percentiles 50 to 75 ranged from 2.07 to 2.83. In conclusion, HOMA index could be a useful tool to detect children and adolescents with the metabolic syndrome. HOMA cut-off values need to be defined in the paediatric population; however, values near to 3 seem to be adequate.