Estimating Mediation Effects in Prevention Studies

描述(由申请人提供)：这项竞争性延续拨款提案的目的是继续开发和评估方法学和统计程序，以确定预防方案如何改变结果变量。这些中介分析评估预防计划所针对的结构上的计划效果和对结果的影响之间的联系。正如许多研究人员指出的那样，中介分析确定了最有效的程序组件，并增加了对导致结果变量变化的潜在过程的理解。在对NIDA资助的几个药物预防项目的分析中，这些新程序的必要性变得显而易见。这笔赠款的首席调查员收到了一笔由NIDA资助的一年期小额赠款，用于研究调解效果的评估，以及一笔为期四年的赠款，用于制定和评估预防研究中常用的设计和程序类型的方法和统计程序。拟议的为期四年的延续研究将审查代表这项研究下一步行动的三个统计专题。每项统计研究都包括分析和模拟研究以及对实际病因学和预防数据的应用。在研究1中，将开发和评估使用重抽样方法和基于两个随机变量乘积分布的方法的显著性检验。这项研究还将对调解的新测试进行评估。研究2将调查项目效果的调解模式，这些模式因参与者的不同而不同。其中一个模型是由治疗调节剂调节的基线模型，其中计划效果取决于中介变量的基线水平。研究3将继续开发和评价潜在增长曲线建模中的调解方法，包括确定具有不同随时间变化轨迹的研究参与者群体或类别的模型。在研究4中，统计方法将应用于NIH资助的几个预防数据集，提供关于模型有用性的重要反馈。研究5将传播有关调解分析的新信息，包括建立调解分析工作组和增加我们的调解分析网站。

DESCRIPTION (provided by applicant): The purpose of this competing continuation grant proposal is to continue developing and evaluating methodological and statistical procedures to determine how prevention programs change outcome variables. These mediation analyses assess the link between program effects on the constructs targeted by a prevention program and effects on the outcome. As noted by many researchers, mediation analyses identify the most effective program components and increase understanding of the underlying processes leading to changing outcome variables. The need for these new procedures became apparent in the analyses of several NIDA funded drug prevention programs. The Principal Investigator of this grant received a one-year NIDA funded small grant to study the assessment of mediation effects and a four-year grant to develop and evaluate methods and statistical procedures for the types of designs and procedures commonly used in prevention research. The proposed four-year continuation study will examine three statistical topics that represent the next steps in this research. Each statistical study includes analytical and simulation research as well as applications to actual etiological and prevention data. In Study 1, significance testing using resampling methods and methods based on the distribution of the product of two random variables will be developed and evaluated. New tests for mediation will also be evaluated in this study. Study 2 will investigate mediation models for program effects that differ by subgroups of participants. One of these models is the mediated baseline by treatment moderator model where program effects depend on baseline levels of a mediating variable. Study 3 will continue the development and evaluation of mediation methods in latent growth curve modeling including models identifying groups or classes of study participants with different trajectories of change over time. In Study 4, the statistical methods will be applied in several NIH funded prevention data sets providing important feedback about the usefulness of the models. Study 5 will disseminate new information about mediation analysis including the development of a mediation analysis working group and additions to our mediation analysis website.