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 involved six main projects. (1) Collaborative work was published on pooling specimens prior to exposure assessment, where pooling is done within levels of a confounder. (2) Another project that is collaborative with Saha-Chaudhuri at McGill in Montreal concerns use of pooling to protect patient confidentiality in a consortium where data need to be provided to an analysis center. (3) Assessment of seasonal effects of the time of year of conception on the risk of preeclampsia applied harmonic analysis via a Cox model to data from the Norwegian Medical Birth Registry, and found evidence for seasonality, i.e. an environmental component in the etiology of preeclampsia, after the statistical analysis was done with careful allowance for confounding. (4) In work that is joint with Shanshan Zhao we are developing improved risk prediction models that account for family structure and breast cancer history and we are applying them to data from the Sister Study on incident breast cancer. (5) Work related to mediation analysis in scenarios where the exposure interacts with a mediator demonstrated that the natural indirect effect is not identifiable when there is exposure/mediator interaction. (6) We are working on a paper related to improving precision by accounting for exposure measurement modifiers. An example is vitamin D, where time of year of blood draw can modify the measurement due to seasonal effects on blood levels.
Showing the most recent 10 out of 20 publications