BACKGROUND: The objectives of the present study were, within a group of 322 healthy European American (EA) and African American (AA) women, to, (1) determine the extent to which insulin sensitivity (Si) was correlated with the elements of the metabolic syndrome using the third Adult Treatment Panel (ATP III) criteria; (2) determine if ethnicity affected the relationships between Si and the elements of the metabolic syndrome; and (3) determine the amount of variance in elements of the metabolic syndrome independently explained by fasting insulin, the acute insulin response to glucose (AIRg), and Si. METHODS: Si and AIRg were assessed with a frequently-sampled intravenous glucose tolerance test and minimal modeling; total body fat with dual-energy x-ray absorptiometry; and intra-abdominal adipose tissue (IAAT) with computed tomography scanning. RESULTS: Among all women combined, Si was associated with fasting glucose (r = -0.18, p < 0.01), waist circumference (r = -0.36, p < 0.001), and high-density lipoprotein cholesteraol (HDL-C; r = 0.18, p < 0.01). However, the association of Si with elements of the metabolic syndrome other than fasting glucose and waist circumference differed with ethnicity; among EA, Si was correlated with triglycerides (r = -0.28, p < 0.01) and HDL-C (r = 0.29, p < 0.001), whereas among AA, Si tended to be correlated with systolic blood pressure (r = -0.17, p = 0.059). In multiple regression modeling, fasting insulin was independently related to more elements of the metabolic syndrome (fasting glucose, waist circumference, triglycerides, systolic and diastolic blood pressure) than were Si or AIRg (both related only to fasting glucose), after adjusting for ethnicity, age, and total fat or IAAT. CONCLUSIONS: Correlation of Si with elements of the metabolic syndrome differed with ethnic group. Fasting insulin, rather than Si, best predicted most elements of the metabolic syndrome.