Statistical Methods In Epidemiology--general

该项目为流行病学开发了新的统计方法，具有广泛的应用，并根据正在进行的流行病学项目，特别是与生殖研究有关的项目，开发了所需的方法。今年的工作涉及三个主要项目。(1)我们正在开发一种经验贝叶斯方法来汇总生育史数据，以预测相同或遗传相关结果的后续风险；(2)我们开发了一种方法来分析怀孕时间研究，这种方法既强大，又避免了由于违反通常的恒定性假设而产生的偏差；以及(3)我们已经开发了一种基于样本池的方法，用于研究病例队列研究中昂贵的测量暴露，这种方法保存了存档的生物标本，并极好地利用了分析预算，但统计能力的损失(甚至是改善)最小。

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 three main projects. (1) We are developing an empirical Bayes approach for summarizing reproductive history data for prediction of subsequent risk of the same or of genetically-related outcomes; (2) We developed methods to analyze time to pregnancy studies in a way that is both powerful and avoids the biases that arise due to violations of the usual constancy assumptions; and (3) We have developed an approach based on specimen pooling for studying expensive-to-measure exposures in case-cohort studies, an approach that conserves archived biospecimens and makes excellent use of the assay budget, but with minimal loss of (or even improvement in) statistical power.