Statistical Issues in AIDS Research

描述（由申请人提供）：在发现艾滋病毒25年后，艾滋病继续流行，目前全世界估计有3300万人感染，每年有250万人新感染。预防和治疗艾滋病毒感染的研究变得越来越复杂，分析的挑战也相应地变得复杂。在本应用程序中，我们概述了发展统计方法的计划，这些方法将直接适用于艾滋病毒/艾滋病研究领域的当前问题。特别是，我们建议开发分析阶梯形随机试验、预防母婴艾滋病毒传播试验、不一致伴侣研究以及采用两期或其他复杂抽样设计的试验或研究的数据的方法。由于HIV感染通常是无症状的，感染时间受两个检测日期的限制，我们继续研究间隔审查数据的方法，并关注与竞争风险和相关间隔审查数据相关的问题。最后，由于寻找有效的HIV疫苗面临越来越大的挑战，我们建议开发新的统计方法来分析基于t细胞的疫苗以及HIV抗原-抗体相互作用的研究数据。公共卫生相关性：本研究将开发用于艾滋病毒感染研究的统计方法。

DESCRIPTION (provided by applicant): Twenty five years after identification of the HIV virus, the AIDS epidemic continues with an estimated 33 million individuals currently infected worldwide and 2.5 million new infections each year. Research to prevent and treat HIV infection has grown increasing sophisticated and the analytic challenges have become correspondingly complex. In this application we outline plans for the development of statistical methods that will be directly applicable to current problems in the field of HIV/AIDS research. In particular, we propose to develop methods for the analysis of data from stepped wedge randomized trials, trials to prevent mother to child transmission of HIV, discordant partner studies, and trials or studies which utilize two-phase or other complex sampling designs. Since HIV infection is typically asymptomatic and the time of infection is bounded by two test dates, we continue work on methods for interval censored data and focus on issues related to competing risks and correlated interval censored data. Finally, as the search for an effective HIV vaccine presents ever greater challenges, we propose to develop novel statistical methods for analyzing data from studies of T-cell based vaccines as well as HIV antigen-antibody interactions. PUBLIC HEALTH RELEVANCE: This research will develop statistical methods for use in studies of HIV infection.