Factor analysis and defining the metabolic syndrome.

OBJECTIVE The metabolic syndrome has been referred to as a number of metabolic or physiologic abnormalities that occur together more often than would be predicted by chance. Considerable controversy exists about the exact abnormalities that are a part of this syndrome. The aim of this study was to examine the interrelations between these abnormalities. DESIGN National Health and Nutrition Examination Survey (1988-1994), a national cross-sectional health survey. SETTING United States. PARTICIPANTS Persons aged > or = 20 years (N=6868). MAIN OUTCOME MEASUREMENTS Factors composed of variables often associated with the metabolic syndrome derived from principal components analysis. RESULTS Depending on the subgroup studied, the analyses suggested that at least 2 or 3 components were needed to explain the majority of variance in a set of variables. Regardless of age group, sex, race or ethnicity, 4 variables (waist circumference, fasting insulin, triglycerides, and high-density lipoprotein cholesterol) consistently loaded together on the first component, which is consistent with a metabolic syndrome factor. Some differences in the number of factors and the loading patterns occurred among 3 age groups and among men and women. Relatively minimal race or ethnic variation was observed when the data were stratified by sex. A subanalysis that included leptin concentrations produced a similar set of factors as the analysis without leptin concentration. Furthermore, leptin concentration did not provide a unifying explanation for the set of factors. CONCLUSIONS Patterns of factors of variables, often associated with the metabolic syndrome, tended to be similar among Whites, African Americans, and Mexican Americans.