STATISTICAL METHODS FOR OUTCOME-DEPENDENT SAMPLING

这项赠款申请的主要目的是开发和评价改进的统计方法，以设计和分析采用结果依赖抽样计划进行的流行病学研究。将研究在特定环境下导致更具成本效益的设计的抽样策略。拟议的方法对癌症和环境研究特别有用，因为暴露错误分类和昂贵的暴露评估是经常面临的挑战。该提案包括四个项目。第一个项目涉及一个具有连续结果变量的一阶段消耗臭氧层物质设计。在这个项目中，将研究两种对回归参数进行推断的方法：对观测到的消耗臭氧层物质样本进行半参数似然方法，以及当结果变量的值对于基本队列已知时，进一步提高效率的伪似然方法。第二个项目考虑了在存在替代和缺失/误测暴露信息的情况下进行病例对照研究的一种新的推断程序。第三个项目涉及一种新的基于半参数似然的推断程序，该程序结合了消耗臭氧层物质样本暴露的较好测量和基础队列总体的较差测量，从而提高了研究效率。第四个项目考虑了一种新的基于半参数似然的推理程序，用于具有连续结果变量的两阶段消耗臭氧层物质设计。本设计利用第一阶段受试者的替代或混杂协变量的信息，在第二阶段的抽样中从特定的子总体中进行目标抽样。将通过理论调查和模拟研究来严格检查每种建议方法的优点和缺点。将与现有方法进行比较。将开发相关软件。将使用开发的方法分析正在进行的关于环境和营养暴露的影响以及对癌症和心血管疾病的影响的流行病学研究的数据集。这些研究包括磁场与乳腺癌风险研究、围产期合作项目、铀矿工的癌症风险研究、社区动脉粥样硬化风险研究、妊娠、感染和营养研究，以及体力消耗、压力和早产的流行病学研究。

The main objective of this grant application is to develop and evaluate improved statistical methods for the design and analysis of epidemiologic studies conducted with an outcome-dependent sampling (ODS) scheme. Sampling strategies that lead to a more cost-effective design in a given setting will be investigated. The proposed methods are particularly useful for cancer and environmental research because exposure misclassification and expensive exposure assessment are frequent challenges. The proposal consists of four projects. The first project deals with a onestage ODS design with a continuous outcome variable. In this project, two methods for making inferences about regression parameters will be studied: a semiparametric likelihood approach for observed ODS sample, and a pseudo-likelihood approach for further improving efficiency when the values of the outcome variable are known for the underlying cohort. The second project considers a new inference procedure for case-control studies in the presence of surrogate and missing/mismeasured exposure information. The third project concerns a new semiparametric likelihood based inference procedure that improves study efficiency by combining a superior measurement of exposure from the ODS sample and an inferior measurement from the underlying cohort population. The fourth project considers a new semiparametric likelihood based inference procedure for a two-stage ODS design with a continuous outcome variable. This design uses the information on surrogate or confounding covariates for subjects in stage one, to target sampling from certain subpopulations in the second stage of sampling. The strengths and weaknesses of each proposed method will be critically examined via theoretical investigations and simulation studies. Comparisons with existing methods will be conducted. Related software will be developed. Data sets from ongoing epidemiologic studies on the effects of environmental and nutritional exposures, and on cancer and cardiovascular disease will be analyzed using the methods developed. These include the Magnetic Fields and Breast Cancer Risk Study, the Collaborative Perinatal Project, the Cancer Risk in Uranium Miners Study, the Atherosclerosis Risk in Communities Study, the Pregnancy, Infection, and Nutrition Study, and the Epidemiology of Exertion, Stress, and Preterm Delivery Study.