Running title: THE METABOLIC SYNDROME IN ADOLESCENTS

Objective: Lack of a universally applicable model for the metabolic syndrome in the pediatric population makes it difficult to define this syndrome and compare its prevalence across studies and diverse populations. We sought to assess whether a single underlying factor could represent the metabolic syndrome in adolescents. Research Design and Methods: Using data from the National Health and Nutrition Examination Survey 1999-2002, we conducted a confirmatory factor analysis to assess the validity of waist circumference, triglycerides, fasting insulin, and systolic blood pressure as potential phenotypic traits for the metabolic syndrome in adolescents aged 12-17 y (n=1262). A multiple-group approach was used to test the invariance in factor loadings across sex and race/ethnicity. Results: The estimates of factor loadings for the total sample were 0.76, 0.46, 0.81, and 0.42 for waist circumference, triglycerides, fasting insulin, and systolic blood pressure, respectively. The goodness-of-fit indices were adequate for the total sample (comparative fit index, 0.99; standardized root mean square residual, 0.02), white boys (1.0; 0.01), African American boys (0.99; 0.03), Mexican American boys (1.0; 0.01), Mexican American girls (1.0; 0.01), and white girls (0.95; 0.04) and acceptable for African American girls (0.94; 0.05). There were no significant differences in factor loadings of the four measured variables between boys and girls and among the three racial or ethnic subgroups. Conclusions: The metabolic syndrome as a single underlying factor for the four simple phenotypic traits may be plausible in adolescents. The proposed model appears to be generalizable across sex and race/ethnicity.