Abstract
BACKGROUND: Metabolic syndrome (MetS) consists of a cluster of cardiometabolic risk factors and is an important determining factor for cardiovascular diseases (CVDs). We intended to use latent class analysis to classify the study population into several clusters. METHODS: The baseline information of 6,814 participants of the Multi-Ethnic Study of Atherosclerosis (MESA) aged 45-84 years in 2000-2002 was used. The latent class analysis was conducted to extract different patterns of components. SAS 9.2 and Stata 12 software were used for analysis. RESULTS: The components of MetS tend to accumulate, hence it would be feasible to categorize the population into three classes: [1] Non-Metabolic Syndrome Latent Class (NonMetS-LC), [2] Low Risk Latent Class (LowR-LC), and [3] Metabolic Syndrome Latent Class (MetS-LC). In women, adding high-density lipoprotein (HDL) component to the two-component combinations of NonMetS-LC will transfer the individual to MetS-LC, and it was found in 100% of combinations of MetS-LC. However, in men, blood pressure (BP) played such a similar role, which was found in 97.36% of combinations of MetS-LC. CONCLUSION: Results showed that clinical value of each MetS component is different by gender. The main component in men was elevated BP; while low HDL and elevated fasting blood sugar (FBS) were in next ranks. However, the main component in women was low HDL; while elevated BP and FBS were in next ranks. Special attention should be paid to BP and HDL components, because these can be useful for clinicians and health policy-makers in diagnosis and screening. In conclusion, this study showed that revisions might be needed for the MetS definitions.