UC San Diego

OBJECTIVES: There were three objectives of the current dissertation: (1) to examine the influence of age and (2) race/ethnicity on the association of adiposity with mortality; and (3) to determine whether leptin or insulin mediate the association of obesity with metabolic, inflammatory, and thrombogenic cardiovascular disease (CVD) risk factors. METHODS: Two population-based cohort studies using data from the third National Health and Nutrition Examination Survey (NHANES III) were used to examine objectives one and two, and a cross-sectional study of NHANES III to examine objective three. NHANES III included a national, complex multistage, clustered, stratified probability sample of the civilian, noninstitutionalized US population. Data were collected at baseline between 1988-1994 during a home interview, physical examination, and laboratory analysis. Mortality follow-up of the baseline cohort was performed through December 2000 with the use of the National Death Index. RESULTS: Age modified the relation of adiposity with mortality. In middle-aged adults (30-64 years), body mass index (BMI) demonstrated a U-shaped relation with mortality in men and a J-shaped relation in women, while positive associations, stronger among women than men, were observed between waist-to-hip ratio (WHR) and mortality. In older adults (65-102 years), decreased adiposity, in general, carried an increased risk for mortality. The association of BMI with mortality varied by sex and race/ ethnicity; however, WHR demonstrated positive relations with mortality, stronger among white and black women than men, but of similar strength among white, black, and Mexican American men. Among Mexican Americans, an elevated WHR increased the risk of death only among US-born English speaking men, while no association was observed in women, regardless of migration, acculturation, or level of education. Insulin demonstrated significant (p<0.05) associations, independent of obesity, with several metabolic CVD risk factors, while leptin was associated only with fibrinogen levels in women. CONCLUSIONS: Age and race/ethnicity are important factors to consider when examining the relation of adiposity with mortality. Insulin may at least partly mediate the association of obesity with metabolic but not inflammatory or thrombogenic CVD risk factors, while leptin does not appear to influence cardiovascular risk through a shared association with CVD risk factors