Statistical Methods In Epidemiology--general

Environmental Health Sciences

Investigators

Linked publications & trials

Abstract

This project develops new statistical methods for epidemiology with broad applications and also methods as needed for ongoing projects in epidemiology, particularly those related to reproductive studies. The work this year focused on computing implications of risk-enriched sampling in a cohort study, with application to the ongoing Sister Study. By studying the first-degree relatives of people who have developed a particular condition, one can build a cohort made up of individuals with higher-than-average risk for that condition, and thereby greatly improve the efficiency of the study. Moreover, such a study will typically include individuals who are well-motivated to provide good data and to remain in the study for a long period of time. The Sister Study is based on building such a cohort, consisting of the sisters of women who have had breast cancer. The accrual rate for incident cases of breast cancer during follow up will consequently be double that of the general population of women in the same age range. This approach enriches for both genetic susceptibility factors and early life environmental factors the two sisters may have shared. In other work related to the theory of directed acyclic graphs, an extension was proposed whereby synergism can be represented, both for pure interaction scenarios (as with PKU) and for statistical interaction depending on scale.