Big Data Digital Outreach and Epidemiology Methods for HIV Care among Communities of Color

摘要 艾滋病毒领域最重要的问题之一涉及解决艾滋病毒护理率低的问题 在艾滋病毒/艾滋病感染者中，特别是在黑人/非裔美国人和拉丁裔社区中。 本申请旨在研究一种新的方法来解决这个问题，通过采用和应用尖端的 顶级技术公司正在使用“大数据推广”方法来提高消费者参与度。 这种方法最近正在取代其他数字推广方法，主要是出于隐私原因， 符合严格的欧盟隐私法-因为它涉及去识别数据。 拟议的数字外联方法已经在卫生（但尚未在艾滋病毒）环境中得到应用。 在COVID-19大流行期间，我们的团队和其他人（包括CDC）研究并发现成功应用 这些方法用于有针对性的数字招募和对COVID-19高风险人群的外展。的结果 COVID-19大流行及其对数字/远程工具使用的影响，这些方法将很快应用于 艾滋病毒，以协助在艾滋病毒研究和护理的目标和参与难以达到的个人。 重要的是，所提出的方法允许访问大规模的被动收集，选择加入，社区和 移动性（GPS ping）数据，已被证明可以添加丰富的粒度数据，以改善健康状况 监督和干预。本申请寻求使用这些新颖的数字外展方法来识别 并招募来自有色人种社区的艾滋病毒/艾滋病感染者，这些人处于高风险之中， 并分析他们的流动性和社区数据，以确定影响艾滋病毒护理的关键地理环境 订婚我们正在为结束艾滋病毒流行（EHE）并与之合作开展这项工作 辖区（华盛顿特区和橙子县卫生部门）和主要参与者利益相关者， 获得他们对需求，实施（包括道德问题）以及未来可能扩大的见解 加强监测和干预工作。具体而言，我们寻求1）识别个人 EHE地区内艾滋病毒/艾滋病感染者/艾滋病患者的高风险，2）使用GPS移动性， 社区（例如，当地犯罪）和艾滋病护理数据，确定影响艾滋病护理的关键地理环境 参与，以及3）与华盛顿特区和橙子县卫生部门合作，探索 对这种方法在个人、人际和 结构层次。 这项为期1年的队列研究将重点确定生活在EHE地区的人， 艾滋病毒护理难以达到，以便与EHE结果措施相一致。据我们所知，这是 第一项将这些新的“大数据外展”方法应用于艾滋病毒的研究，将招募迄今为止最大的队列， GPS移动性，社区和其他艾滋病毒护理背景数据，以及第一项被动收集的艾滋病毒研究 流动性数据，与以前的研究相比，这有助于提高数据质量和降低辍学率。

Abstract One of the most significant problems in the field of HIV deals with addressing the low rates of HIV care among individuals living with HIV/AIDS, especially among Black/African American and Latinx communities. This application seeks to study a novel way to address that problem by adopting and applying a cutting-edge “big data outreach” approach being used to increase consumer engagement by top technologies companies. This approach has recently been replacing other digital outreach methods largely for privacy reasons-- to conform to stringent European Union privacy laws-- as it involves de-identified data. The digital outreach method being proposed is already being applied in health (but not yet HIV) settings. During the COVID-19 pandemic, our team and others (including the CDC) studied and found success applying these methods for targeted digital recruitment and outreach to those at high-risk for COVID-19. As a result of the COVID-19 pandemic and its effect on use of digital/remote tools, these approaches will soon be applied to HIV to assist in targeting and engaging hard-to reach individuals in HIV research and care. Importantly, the proposed methods allow access to large-scale passively-collected, opt-in, community and mobility (GPS pings) data, which have been shown to add rich and granular data to improve health surveillance and interventions. This application seeks to use these novel digital outreach methods to identify and enroll individuals living with HIV/AIDS from communities of color who are at high-risk for being out of care, and analyze their mobility and community data to identify the key geographic contexts that impact HIV care engagement. We are conducting this effort for, and in partnership with, 2 Ending the HIV Epidemic (EHE) jurisdictions (Washington D.C. and Orange County health departments) and key participant stakeholders to gain their insights on needs, implementation (including ethical concerns), and potential future scale-up of this approach to improve surveillance and intervention efforts. Specifically, we seek to 1) Identify individuals of color living with HIV/AIDS within EHE regions who are at high-risk for being out of care, 2) Using GPS mobility, community (e.g., local crime), and HIV care data, identify the key geographic contexts that impact HIV care engagement, and 3) In partnership with the Washington D.C. and Orange County health departments, explore a case study of the ongoing barriers and facilitators of this approach at the individual, interpersonal, and structural levels. This 1-year cohort study will be focused on identifying people living within an EHE region who have been hard-to-reach for HIV care in order to converge with EHE outcome measures. To our knowledge, this is the first study to apply these novel “big data outreach” methods to HIV, will enroll the largest cohort to date with GPS mobility, community, and other HIV care contextual data, and the first HIV study to passively collect mobility data, which helps to increase data quality and reduce dropout rates compared to previous studies.