Optimizing outcomes at the intersection of HIV and mental health: Prediction, precision medicine, and population health

项目总结 精神障碍和药物滥用在艾滋病毒携带者(PWH)和 对艾滋病毒控制的负面影响。这些情况造成了巨大的情感和身体负担，阻碍了 病毒学抑制的成就，并与增加风险的行为有关 变速箱。虽然PWH有多种精神疾病诊断是很常见的，但人们对此知之甚少 关于精神症状和药物滥用行为相结合的方式 艾滋病毒控制不良的风险增加，或者如何个性化管理这些常见的共生现象 精神错乱。此外，对精神卫生服务创新对公共健康的潜在影响的估计 艾滋病毒的流行通常是缺乏的。统计学和机器学习方面的最新创新使之成为可能 利用复杂的数据来确定威尔斯亲王医院的精神疾病模式，并为其量身定做精神疗法 以传统统计方法无法做到的方式优化艾滋病毒控制。 应聘者是翻译和计算调查员和普通内科医生，网址为 拥有计算机科学和生物统计学背景的约翰霍普金斯大学。在本次颁奖期间， 他将得到一个团队的指导，该团队的专业知识涵盖艾滋病毒护理、精神流行病学、机器学习、 人口层面的艾滋病毒建模，以及个性化的药物。候选人的长期职业目标是成为 将创新的统计和机器学习方法应用于 改善处于艾滋病毒和心理健康交汇点的个人和人群的健康。 该项目的总体目标是解决精神障碍和药物滥用问题 通过开发利用复杂数据为患者护理提供信息的模型，了解艾滋病毒连续护理的背景 和公共政策。将实现三个目标：(1)确定抑郁症的症状组合 焦虑和滥用酒精、可卡因、类阿片、大麻和安非他明会影响个体水平的艾滋病毒 对照；(2)使用重复的患者报告的抑郁、焦虑和物质滥用的测量来预测 未来在潜在药物疗法下的精神健康和艾滋病毒控制；以及(3)发展人口 预测精神卫生服务干预措施对艾滋病毒发病率和死亡率的影响的模型。这个 模型将利用艾滋病中心综合临床系统研究网络的数据 (CNICs)，一个种族多元化的多地点队列，超过31,000名威斯康星医院。 这项有指导的研究将伴随着心理健康流行病学的相关技能培训， 关注威斯康星医院，以及先进的贝叶斯和机器学习方法。总的来说，这项研究和职业生涯 发展培训将提供一条通往独立职业生涯的明确途径，成为一名专注于临床研究的 威尔斯亲王医院精神障碍和药物滥用的个体优化管理 和人口水平。

PROJECT SUMMARY Psychiatric disorders and substance misuse are prevalent among people with HIV (PWH) and have negative effects on HIV control. These conditions impose substantial emotional and physical burdens, impede the achievement of virologic suppression, and are associated with behaviors that increase the risk of transmission. While it is common for PWH to have multiple psychiatric diagnoses, relatively little is known about the ways in which combinations of psychiatric symptoms and substance misuse behaviors confer elevated risk for poor HIV control, or how to personalize the management of these commonly co-occurring disorders. Furthermore, estimates of the potential public-health impact of mental health service innovations on the HIV epidemic are generally lacking. Recent innovations in statistics and machine learning make it possible to harness complex data to identify patterns of psychiatric disease in PWH and tailor psychiatric therapy to optimize HIV control in ways that classical statistical approaches cannot. The candidate is a translational and computational investigator and general internal medicine physician at Johns Hopkins University with a background in computer science and biostatistics. During this award period, he will be mentored by a team whose expertise spans HIV care, psychiatric epidemiology, machine learning, population-level HIV modeling, and personalized medicine. The candidate's long-term career goal is to become an independent, translational researcher who applies innovative statistical and machine learning approaches to improve the health of individuals and populations at the intersection of HIV and mental health. The overarching objective for this project is to address psychiatric disorders and substance misuse in the context of the HIV continuum of care by developing models that leverage complex data to inform patient care and public policy. Three aims will be undertaken: (1) to identify how combinations of symptoms of depression and anxiety and misuse of alcohol, cocaine, opioids, marijuana, and amphetamines impact individual-level HIV control; (2) to use repeated patient-reported measures of depression, anxiety, and substance misuse to predict future mental health and HIV control under a potential pharmacotherapies; and (3) to develop population models that project the impact of mental health service interventions on HIV incidence and mortality. The models will leverage data from the Centers for AIDS Research Network of Integrated Clinical Systems (CNICS), a racially diverse, multi-site cohort of over 31,000 PWH. This mentored research will be accompanied by relevant skills training in mental health epidemiology, the care of PWH, and advanced Bayesian and machine-learning methods. Collectively, this research and career development training will provide a clear pathway to an independent career as a clinical investigator focused on optimizing the management of psychiatric disorders and substance misuse in PWH at both the individual and population level.