| Early stages of chronic kidney disease (CKD) have been recognized not only as precursors to end stage renal disease (ESRD), but also as predictors of cardiovascular disease and mortality. The social epidemiology of CKD is not well developed. This dissertation uses the Life Course Socioeconomic Status (LCSES) ancillary study of the Atherosclerosis Risk in Communities Study (ARIC) to describe the associations between individual social status, the socioeconomic characteristics of the areas subjects have lived in, and CKD. Social class was dichotomized as worker vs. non-worker for subjects' fathers and for subjects at ages 30, 40, and 50 using data collected on workplaces roles and ownership; education attainment was also used. Area socioeconomic status (SES) was defined by a composite score of standardized census characteristics that were weighted using principle components analysis (PCA) of childhood county of residence and of census tracts at ages 30, 40, and 50; subjects were then placed into race-specific tertiles of area SES. CKD was defined by glomerular filtration rate (GFR) below 45 ml/min, or by ICD-9 discharge diagnosis codes. Three models of life course socioeconomic position were used: life course periods; cumulative exposure; and class trajectories. The life course model employed multilevel analysis to measure the association of CKD with social class and area SES at each of the four periods in the life course. All analyses were stratified by race, and missing data were imputed. Using the life course periods model, noteworthy associations were found for White and African American subjects who were working class at age 30; among both races, area SES was associated with CKD at age 50. Associations between education level and CKD were weak. The cumulative and trajectory class models both showed the strongest associations for being working class across the life course. Cumulative exposure to low area SES does not appear to be related to CKD, but a downward trajectory of area-level SES was related to the outcome. Associations between socioeconomic characteristics and CKD were stronger among African Americans than Whites. Many of these associations remained after adjusting for the two most common determinants of CKD, diabetes and hypertension. |
