Pregnancy and response to antiretroviral therapy in South Africa

描述（由申请人提供）：本提案的总体目标是支持首席研究员Daniel Westreich博士成为撒哈拉以南非洲艾滋病毒阳性妇女怀孕研究的主要健康科学家，特别关注怀孕对南非高效抗逆转录病毒治疗（HAART）反应的影响。南非孕妇中艾滋病毒的流行率大大高于其他成年人。虽然在怀孕期间和怀孕后对抗逆转录病毒治疗的反应涉及预防母婴传播艾滋病毒的问题，但对怀孕本身对治疗反应的影响知之甚少。了解怀孕对产妇对HAART治疗反应的影响，对于向非洲和其他地区的艾滋病毒阳性妇女提供良好的临床护理至关重要。为了解决这些问题，本研究建议检查怀孕（治疗开始时怀孕和治疗开始后怀孕）对HAART的病毒学反应、坚持HAART、保留护理、对HAART的免疫反应和临床（艾滋病/死亡）对HAART的反应的影响。由于不可能（而且非常不道德）随机化怀孕暴露，本研究将依赖于南非约翰内斯堡Themba Lethu诊所收集的高质量临床数据，该诊所是一家与威特沃特斯兰德大学临床艾滋病毒研究部门相关的城市抗逆转录病毒治疗诊所。拟议工作的首席研究员Daniel Westreich博士在与诊所的研究人员和临床医生合作进行高质量的研究方面有着良好的记录。对这些数据的分析将需要使用结构模型，这是一种先进的流行病学方法。因此，在本奖项的指导阶段，Westreich博士将与导师Stephen R. Cole博士（国际公认的结构模型权威）合作，获得这些技能，并评估怀孕对产妇对HAART病毒反应的影响。本次培训将在北卡罗来纳大学教堂山分校流行病学系进行，这是一个国际知名的流行病学和公共卫生研究卓越中心。在这一培训和经验的基础上，在该奖项的独立阶段，Westreich博士将领导调查，评估怀孕对保留护理和坚持HAART治疗的影响，以及对HAART治疗的免疫和临床反应的影响。这项工作将回答一系列关键的实质性问题，围绕怀孕对南非对HAART的反应的影响。这样做，这项工作将大大增加关于这一重要问题的科学和公共卫生证据的基础，很可能以一种可推广到撒哈拉以南非洲大部分地区的方式。此外，这项工作和培训将帮助Westreich博士发展成为一名健康科学家，使他在应用结构模型和其他流行病学方法对复杂的纵向数据进行定量因果推理方面具有专业知识。

DESCRIPTION (provided by applicant): The overall goal of this proposal is to support the principal investigator, Dr. Daniel Westreich, in becoming a leading health scientist in the study of pregnancy among HIV+ women in sub-Saharan Africa, with a particular focus on the effect of pregnancy on response to highly active antiretroviral therapy (HAART) in South Africa. Prevalence of HIV among pregnant women in South Africa is substantially higher than that among other adults. While responses to antiretroviral therapy during and after pregnancy with respect to issues of prevention of mother to child transmission of HIV, little is known about the effect of pregnancy itself on responses to therapy. Understanding the effect of pregnancy on maternal responses to HAART is essential to providing good clinical care to HIV+ women in Africa and elsewhere. To address these questions, this work proposes to examine the effect of pregnancy (both pregnancy present at time of initiation of therapy, and pregnancy which occurs after the initiation of therapy) on virologic responses to HAART, adherence to HAART, retention in care, immunologic responses to HAART, and clinical (AIDS/death) responses to HAART. Because it is impossible (and highly unethical) to randomize the exposure of pregnancy, this study will rely on high quality clinical data collected in the Themba Lethu Clinic in Johannesburg, South Africa, an urban antiretroviral therapy clinic associated with the Clinical HIV Research Unit at the University of the Witwatersrand. The Principal Investigator of the proposed work, Dr. Daniel Westreich, has a strong track record in producing high quality research in collaboration with investigators and clinicians at the Clinic. The analysis of these data will require the use of structural models, an advanced body of epidemiologic methods. Thus, during the mentored phase of this award, Dr. Westreich will work with Mentor Dr. Stephen R. Cole, an internationally recognized authority on structural models, to acquire these skills and to evaluate the effect of pregnancy on maternal virologic responses to HAART. This training will take place in the Department of Epidemiology at The University of North Carolina at Chapel Hill, an internationally renowned center of excellence in epidemiologic and public health research. Building on this training and experience, during the independent phase of the award Dr. Westreich will lead investigations to evaluate the impact of pregnancy on retention in care and adherence to HAART, and on immunologic and clinical responses to HAART. This work will answer a suite of key substantive questions around the impact of pregnancy on response to HAART in South Africa. In doing so, this work will dramatically increase the base of scientific and public health evidence on this important subject, most likely in a way which is generalizable to much of sub-Saharan Africa. In addition, this work and training will help Dr. Westreich develop as a health scientist, giving him expertise in the application of structural models and other epidemiologic methods for quantitative causal inference in complex longitudinal data.