?Mediation and moderation of the effects of obesity on coronary heart disease in

Univ Of North Carolina Chapel Hill, Chapel Hill NC

Investigators

Abstract

[unreadable] DESCRIPTION (provided by applicant): Excess adiposity is associated with coronary heart disease (CHD), high blood pressure, elevated glucose and dyslipidemia. Racial and gender differences have been found in the prevalences of obesity, CHD risk factors and CHD outcomes, as well as in the associations among these variables. Few studies have examined mediation of the obesity-CHD event relationship by CHD risk factors or compared mediation across risk factors, and it is unknown whether risk factors mediate the obesity-CHD relationship to the same degree in African Americans and Whites or in men and women. The lack of such studies is due in part to the scarcity of appropriate data and in part to limitations of traditional epidemiologic methods. Mediation is commonly analyzed in other fields using structural equation modeling (SEM). SEM has several advantages over standard regression for mediation, including specification of each path (and non-path) among variables and modeling of measurement error. SEM also permits significance-testing of mediated effects. Aim 1. Examine risk factors as mediators of the relationship between obesity (body mass index (BMI) and waist circumference) and time to CHD. SEM will be used with Cox proportional hazards regression to assess mediation by systolic and diastolic blood pressure, glucose, LDL and HDL cholesterol and triglycerides. 1.a. Examine and compare metabolic risk factors individually as mediators of the effect of obesity on CHD 1.b. Examine and compare metabolic risk factors in combination as mediators of the effect of obesity on CHD 1.c. Determine if mediation of the obesity-CHD relationship differs by race and/or gender A comparison of effects of obesity on CHD through blood pressure, glucose and lipids must consider how to incorporate medication use in the analysis. This issue is especially important for blood pressure medication because of the high prevalence of antihypertensive medication use among middle-aged Americans. Aim 2. Investigate the effect of medication use on relationships of BMI and waist circumference with blood pressure and CHD. This secondary aim will be used to inform the analyses for the primary aim, above. 2.a. Determine if antihypertensive medications modify the cross-sectional associations between obesity and blood pressure 2.b. Explore how antihypertensive, lipid-lowering and diabetes medications affect mediation of the relationship between obesity and time to CHD These analyses will use data from the Atherosclerosis Risk in Communities (ARIC) Study, a cohort of 15,689 African American and White adults aged 45-64 at baseline (1987-1989). Follow-up data on fatal and non-fatal CHD events through 2004 will be available. To our knowledge, this will be the first study to use SEM to examine and compare metabolic risk factors as mediators of the adiposity-CHD relationship, as well as the first to systematically examine differences in mediation across race-gender groups. [unreadable] [unreadable] [unreadable]