We propose to continue our studies of genetic and nongenetic influences on several recently identified cardiovascular disease risk factors using data from the NHLBI Twin Study cohort. This unique database includes data from the 134 monozygotic and 134 dizygotic twin pairs who participated in the third examination. At this examination, plasma insulin levels, apolipoprotein A-I and B, and HDL subfractions were measured. These will be the primary outcome variables in the proposed analyses. The investigators participated in the collection of third examination data (in 1986-1987) and have been conducting analyses in this database for the past three years. The analyses proposed herein are a continuation of this research program. The unique characteristics of a twin database allow calculation of heritability estimates as measures of genetic effects, both before and after adjusting for potential environmental or behavioral influences such as obesity. Matched monozygotic co-twin analyses, using linear regression techniques, provide a method for determining whether the relationship between a dependent variable and a set of predictors persists after controlling for potential genetic confounding (i.e., whether nongenetic variation is related to the outcome of interest. Specifically, we will estimate the heritability of the postload plasma insulin level before and after adjustment for level of obesity (body mass index). Next, using the matched co-twin method, we will examine the relation of nongenetic (i.e. behavioral) variation in obesity and body fat distribution, and of other behavioral factors including leisure-time physical activity, alcohol consumption and cigarette smoking, with postload plasma insulin. Heritability estimates have already been reported for apolipoproteins and HDL subfractions, but we will again examine the influence of nongenetic variation in obesity and fat distribution and of other behavioral factors on these cardiovascular disease risk factors. Demonstration that nongenetic variation in obesity is related to insulin, apolipoprotein levels or HDL cholesterol subfractions would provide additional support for intervention programs that attempt to prevent or reduce obesity.
