New Methods for the design and evaluation of large HIV prevention interventions

摘要 新发现的预防艾滋病毒的有效工具的出现， 新的挑战和对新范式的需求：我们如何评估 当我们有证据表明， 他们各自致力于减少艾滋病毒的传播？我们如何推出有效的 战略，但评估其有效性和成本效益在现实世界的设置， 规模？在四个主要的艾滋病毒预防机构之间的有力合作中， 目前正在进行的干预措施-包括哈佛的博茨瓦纳社区预防 项目和家庭萨拉马研究的PMTCT，范德比尔特和帝国的PopART项目， 和非洲中心的治疗即预防试验，我们的建议，题为“新的 设计和评估大型艾滋病毒预防干预措施的方法”将 解决现有设计和分析方法中的关键差距， 大规模艾滋病预防干预项目。在与该组织进行广泛对话后， 这些项目的主要研究人员和其他关键人员（其中三人是 我们确定了我们的具体目标：1）发展 并验证了新的网络方法，以改善大型组合的评估 2）发展先进的统计方法，以改善预防临床试验 设计和评估大规模的艾滋病预防研究。3)加强预防 试验设计，以调整因失访导致的偏倚，并评估传染性， 间接影响。4)理清并评估特定干预措施的效果， 包括哈佛-博茨瓦纳试验和HPTN 061/PopART试验中的“包装"，

Abstract The advent of newly discovered efficacious tools for the prevention of HIV creates a new challenge and the need for new paradigms: How do we assess the impact of combinations of HIV prevention tools in ethical ways, when we have evidence that individually they work to reduce HIV transmission? How do we roll out efficacious strategies, but assess their effectiveness and cost-effectiveness in real-world settings at scale? In a powerful collaboration between four of the major HIV preventive interventions currently underway - including Harvard's Botswana Community Prevention Project and Familia Salama study of PMTCT, Vanderbilt and Imperial's PopART project, and the Africa Centre's Treatment as Prevention trial, our proposal, entitled "New methods for the design and evaluation of large HIV prevention interventions" will address key gaps in the available methodologies for the design and analysis of these large- scale HIV preventive intervention projects. After extensive dialogue with the Principal Investigators and other key personnel on these projects (three of whom are Co-investigators of this proposal), we have identified our specific aims as: 1) To develop and validate novel network methods to improve the evaluation of large combination prevention clinical trials 2) To develop advanced statistical methods to improve the design and evaluation of large- scale HIV prevention studies. 3) To augment prevention trial designs to adjust for bias due to loss to follow-up and to assess infectiousness and indirect effects. 4) To disentangle and evaluate the effects of specific interventions that comprise a 'package' as in the Harvard-Botswana and HPTN 061/PopART trials