IHBEM: Modeling the influence of individual and network behavior on HIV/HCV transmission dynamics

在世界各地，政府和社区都在努力控制艾滋病毒/艾滋病的传播。然而，尽管取得了巨大进展，但仍有一些人难以控制流行病。其中一个群体是注射吸毒者（PWID）。在撒哈拉以南非洲以外，大约五分之一的新艾滋病毒感染者是PWID患者，而在美国，截至2018年底，PWID患者人数估计为370万。丙型肝炎（HCV）是对PWID造成不成比例影响的另一种病毒。为了消除或控制艾滋病毒和丙型肝炎病毒，重要的是要了解驱动传播的因素，并了解如何最好地进行干预，以防止病毒的传播。由于PWID患者的社会网络复杂，包括他们的注射伙伴和他们生活的社区，这些因素很难在PWID患者中确定。这项工作的目标是改进和开发干预措施，以控制艾滋病毒和丙型肝炎病毒在PWID中的传播。该项目将行为、社会经济和流动性数据与传染病传播动态相结合。跨学科的研究团队汇集了流行病学、临床医学、行为科学、工程学和数学建模。该项目利用大型国际数据集开发复杂的传输模型，然后使用机器学习和其他新颖的数据科学技术来填补缺失的数据并训练模型。这些模型可以适用于任何社区，以了解当地PWID社区中驱动HIV和HCV传播的因素。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

All around the world, governments and communities have worked to control the spread of HIV/AIDS. However, despite dramatic progress, there are populations of individuals in which epidemic control has been difficult. One of these populations is composed of people who inject drugs (PWID). About one in five new HIV infections outside of Sub-Sharan Africa are in PWID, and in the US the number of PWID was estimated to be 3.7 million at the end of 2018. Hepatitis C (HCV) is another virus that has had a disproportionate impact on PWID. To eliminate or control HIV and HCV it is important to understand the factors that drive transmission and learn how best to intervene to prevent transmission of the viruses. These factors are difficult to identify among PWID because of the complex social networks of PWID, including their injection partners and the communities in which they live.The goal of this work is to improve and develop interventions to control the spead of HIV and HCV in PWID. This project integrates behavioral, socioeconomic and mobility data with infectious disease transmission dynamics. The interdisciplinary research team brings together epidemiology, clinical medicine, behavioral science, engineering, and mathematical modeling. The project leverages large international datasets to develop complex transmission models and then uses machine learning and other novel data science techniques to fill in missing data and train the models. These models can be adapted to be used in any community to understand the factors that drive HIV and HCV transmission in their local PWID community.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.