Understanding HIV Prevalence and Modeling the Impact of PrEP on HIV Incidence among Transgender and Non-binary Adults in Seattle, Washington

摘要 在美国，变性人和非二进制(TNB)人感染艾滋病毒的风险很高。艾滋病毒流行率为14% 在所有TNB成年人中，黑人变性人女性的比例高达44%。由于结构性障碍，耻辱， 由于担心受到虐待，TNB患者不太可能从事包括艾滋病毒预防和护理在内的医疗保健。 因此，在TNB社区中，PrEP摄取率、ART覆盖率和病毒抑制率较低，这些因素 也增加了艾滋病毒感染和传播的风险。我们迫切需要解决这个被忽视的高风险 通过提高我们对TNB人群中艾滋病毒流行病学的了解，以及通过制定 艾滋病毒防治的有针对性的干预措施。该提案涉及以下问题：谁 TNB的性伴侣是成年人吗？他们的性网络与其他人群有何重叠？多么 TNB成人中的艾滋病毒流行率在美国各地和不同种族/性别亚群之间是否存在差异？其影响是什么？ 增加的PrEP覆盖率？我们将使用三种方法来回答这些问题：三个三角测量 来自华盛顿州西雅图的关于艾滋病毒危险行为、小区域估计和艾滋病毒数学模型的数据集 变速箱。西雅图已被确定为48个县之一，超过50%的新艾滋病毒诊断发生在这些县。 因此，这项研究将解决关于西雅图和全国TNB人员的关键知识差距，以提供信息 地方和国家艾滋病毒预防和公共卫生活动。目标1：描述TNB的HIV危险行为 华盛顿州西雅图TNB人群的个体和性伴侣的横断面数据 独立的本地数据来源：(1)西雅图的性病诊所，(2)公共卫生-西雅图和金县的骄傲 (3)全国艾滋病毒行为监测(NHBS)调查。目标2：估计疾病的患病率 在居住在大都市统计区域的美国所有县的TNB个人中，使用Small 面积估算方法。小面积估计描述了一组用于计算 很少观测的小地理区域的流行率，直接估计无法提供 由于样本量小，统计数据可靠。我们将应用贝叶斯分层小区域估计模型 根据2015年美国变性人调查的全国数据，这是迄今为止最大的全国TNB成年人样本， 超过27,000名受访者。目的3：建立HIV在TNB成人中传播的流行模型 在西雅图，华盛顿州预测PrEP使用量的增加如何影响变性人女性10年内的艾滋病毒发病率， 变性人和非双性恋的成年人。这项研究将为临床护理和有针对性的公共卫生提供信息 对性别少数群体及其性伴侣的干预。县级艾滋病毒估算值的发展 在美国的流行将支持对TNB人群中微流行病的理解。 数学模型是预测有针对性的干预措施对艾滋病毒发病率影响的有力工具， 在缺乏TNB人群干预措施的纵向数据的情况下，这一点很重要。几个流行病模型 包括TNB个人，Aim 3将第一个包括变性人和非双性恋者。

ABSTRACT Transgender and non-binary (TNB) people in the U.S. are at high risk for HIV acquisition. HIV prevalence is 14% among all TNB adults, and as high as 44% among black transgender women. Due to structural barriers, stigma, and fear of mistreatment, TNB people are less likely to engage in health care, including HIV prevention and care. Consequently, PrEP uptake, ART coverage, and viral suppression are low in the TNB community, factors which also increase HIV acquisition and transmission risk. We urgently need to address this neglected high risk population by improving our understanding of the epidemiology of HIV in TNB populations and by developing of targeted interventions for HIV prevention and treatment. This proposal addresses the following questions: Who are the sex partners of TNB adults, and how does their sexual network overlap with other populations? How does HIV prevalence among TNB adults vary across the US and by racial/gender subgroups? What is the impact of increased PrEP coverage? We will use three approaches to answer these questions: triangulation of three datasets from Seattle, WA on HIV risk behaviors, small area estimation, and mathematical modeling of HIV transmission. Seattle has been identified as one of 48 counties in which over 50% of new HIV diagnoses occur. Thus, this research will address critical knowledge gaps about TNB people in Seattle and nationally to inform local and national HIV prevention and public health activities. AIM 1: Characterize HIV risk behaviors of TNB individuals and sexual partners of TNB persons in Seattle, WA using cross-sectional data from three independent local data sources: (1) Seattle's STD Clinic, (2) Public Health-Seattle & King County's Pride Surveys, and (3) the National HIV Behavioral Surveillance (NHBS) survey. AIM 2: Estimate the prevalence of HIV among TNB individuals for all U.S. counties residing in metropolitan statistical areas using small area estimation methods. Small area estimation describes a collection of statistical methods used to calculate prevalence for small geographic areas with few observations, for which direct estimates would not provide reliable statistics due to small sample sizes. We will apply a Bayesian hierarchical small area estimation model to national data from the 2015 U.S. Transgender Survey, the largest national sample of TNB adults to date, with over 27,000 respondents. AIM 3: Develop an epidemic model of HIV transmission among TNB adults living in Seattle, WA to predict how increased PrEP use impacts 10-year HIV incidence for transgender women, transgender men, and non-binary adults. This research will inform clinical care and targeted public health interventions for gender minorities and their sex partners. The development of county-level estimates of HIV prevalence across the U.S. will support the understanding of microepidemics among TNB populations. Mathematical models are powerful tools for predicting the impact of targeted interventions on HIV incidence, and are important in the absence of longitudinal data on interventions for TNB populations. Few epidemic models include TNB individuals, and Aim 3 will be the first to include transgender men and non-binary individuals.