from application) The Cache County, Utah, Study on Memory in Aging, a longitudinal study of dementia and cognitive decline, is well-suited for development of longitudinal epidemiologic studies of genetic and environmental determinants of age-specified survival traits and rate-of-change traits. We propose the following specific aims to establish the Cache County Family-based Cohort Study on Aging: 1. Data from the original 5092 Cache cohort members will be linked to the computerized genealogical records of the Utah Population Data Base. The family structure of each cohort member will be analyzed and variables will be derived for each participant's average familial longevity and familial late female fertility. We will examine the associations between familial longevity and late female fertility and age-specified traits. 2. The estimated 4100 surviving cohort members will be re-contacted and interviewed to collect data on their health history and the vital status and contact information for their relatives. Blood samples will be obtained, processed, and banked for future genetic, nutritional, and biomarker studies. 3. Data on the relatives ascertained in Aim 2, including an estimated 11,000 living siblings, will be linked to the pedigree database of original cohort members using the Utah Population Data Base. In a pilot study we will select a random sample of the living siblings living outside of Cache County and recruit 400 of them for data collection in order to determine the feasibility of full-scale family-based association studies that extend beyond Cache County and beyond Utah. 4. New statistical tools will be developed for the study of age-specified traits in family-based association studies. We propose to extend the unified modeling strategies for family-based association studies in three new areas: a) application in studies with the absence of parental genotype data; b) solution of the problem of handling the nuisance mean of phenotypic distributions; and c) consideration of phenotype as a censored survival outcome. We will investigate use of the unified modeling approach--within the sib-pair framework--to test the degree of association between candidate genes and age-specified traits and to test for gene-gene and gene-environment interactions.
Showing the most recent 10 out of 19 publications
