The use of existing cohort studies provides an opportunity to infer genetic predisposition to late-onset disease. One difficulty with using data collected from ongoing cohorts is that DNA sampling may have occurred substantially after study inception, so that a nontrivial fraction of subjects may have been censored prior to genotyping. This project applies a statistical modeling framework that accounts for censored information to data gathered on the original cohort of the Framingham Heart Study to measure the effect of different Apolipoprotein-E (Apo- E) genotypes on the age of occurrence of various cardiovascular disease events. Specifically, the result of model fitting will address the following specific aims: 1. To measure the effect of different Apo-E genotypes on the age of cardiovascular disease onset, 2. To measure the presence and magnitude of an interaction of Apo-E genotypes with gender on age of cardiovascular disease and onset 3. To measure the distribution of Apo-E polymorphism in the study population, accounting for missing genetic information due to censoring. The long-term goals of this project are to use the developed model components as the basis for more complex analyses of studies on cardiovascular disease events. For example, larger data sets where pedigree information may be available (such as the offspring cohort in the Framingham Heart Study) may serve as a more complete study of genetic effect on cardiovascular disease. The benefit of this project to public health is that establishing the genetic contribution to disease onset has important implications for genetic and medical counselling, as well as further awareness of the importance of genetic profiling in medical research.
| Glickman, Mark E; Kao, Mei-Fang (2005) Apo-E genotypes and cardiovascular diseases: a sensitivity study using cross-validatory criteria. Biom J 47:541-53 |
