Collaborative Research: Modeling the opioid and HIV epidemics

在过去的15年里，美国因阿片类药物流行而死亡的人数增加了四倍。阿片类药物使用量的增加，特别是注射毒品使用量的增加，直接和间接地助长了艾滋病毒等传染病的传播。吸毒还会降低抑制力，进而促进更多的艾滋病毒传播。这导致美国艾滋病毒病例下降的趋势趋于平稳。该项目的目标是开发新的、数据驱动的、多尺度的阿片类药物和艾滋病毒流行相互交织的模型。这些模型将有助于更好地了解流行病之间的相互作用，并可用于指导公共卫生政策，例如扭转艾滋病毒病例的近期趋势。这些模型将使用美国艾滋病毒和阿片类药物使用的数据，以及艾滋病毒携带者和吸毒者的病毒血症数据。考虑这两种流行病的动态相互作用将提供人们单独关注其中一种时所缺失的洞察力，包括个人内部的过程如何影响人口规模的动态。我们的初步结果已经确定了有希望的控制方法，即以药物滥用流行为目标可以降低吸毒者感染艾滋病毒的风险。PI将测试这些控制策略在多个建模场景中的稳健性。这项研究将有助于确定控制这两种相互交织的流行病的改进策略。为了了解这两种流行病，将建立阿片类药物和艾滋病毒流行的多尺度免疫流行病学模型，并进行分析和模拟。这一模型明确包括艾滋病毒在宿主体内的动态以及这种动态对宿主间艾滋病毒传播的影响，以及阿片类药物使用对传染病动态各个层面的影响。该模型将与多尺度数据相联系，并将进行可辨识性分析。将开发一种新的多尺度模型的网络版本，允许对具有不同联系人数量的个人进行建模。将利用即将开发的新的分析技术，从分析和数字两个方面研究接触网络结构和宿主内部动态对两种相互关联的流行病的综合影响。最后，将最优控制方法应用于多尺度免疫流行病学网络模型，以表征控制策略的效果。为了研究初步结果中确定的两种控制方法(针对药物滥用和降低吸毒者的艾滋病毒风险)的稳健性，将这些方法与最优控制模型的结果进行比较。多尺度数据和模拟的严格结合将揭示采用建模框架的最佳控制策略的任何依赖关系。所有这些结果将综合在一本关于免疫流行病学建模的书中，该书结合了这项研究的要素，在这一快速发展的领域的广泛视角下。这本书将对研究生、学者和实践者有用。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

In the last 15 years, US deaths due to the opioid epidemic have quadrupled. The increase in opioid use, and in particular, in injection drug use, has both directly and indirectly contributed to the spread of infectious diseases such as HIV. Drug use also lowers inhibitions, which in turn facilitates more HIV transmission. This has resulted in a leveling off of the decrease of HIV cases in the US. The goal of this project is to develop novel, data-driven, multi-scale models of the intertwined opioid and HIV epidemics. Such models will lead to a better understanding of the interplay of the epidemics and can be used to guide public health policy, for instance to reverse the recent trends in HIV cases. The models will use data on HIV and opioid use in the US as well as viremia data for individuals who both have HIV and use drugs. Considering the dynamic interplay of the two epidemics will provide insights that are missing when one focuses on either alone, including how processes within individuals bear on dynamics at the population scale. Our preliminary results have already identified promising methods of control that targeting the drug abuse epidemic could reduce HIV risk among drug users. The PIs will test the robustness of these control strategies in multiple modeling scenarios. This research will help identify improved strategies to control these two intertwined epidemics.To understand the two epidemics, a multi-scale immuno-epidemiological model of the opioid and HIV epidemics will be developed, analyzed and used for simulations. This model explicitly includes the dynamics of HIV within hosts and the effect of such dynamics on transmission of HIV between hosts, as well as the effect of opioid use on each level of infectious disease dynamics. The model will be linked to multi-scale data, and identifiability analyses will be performed. A novel network version of the multi-scale model will be developed that allows modeling of individuals with varying number of contacts. The impact of the combined effects of contact network structure and within-host dynamics on the two linked epidemics will be investigated both analytically and numerically, using novel analytical techniques that will be developed. Finally, optimal control approaches will be applied to the multi-scale immuno-epidemiological network model in order to characterize effect of control strategies. To study the robustness of the two control methods identified in the preliminary results (targeting drug abuse and reducing HIV risk among drug users), a comparison of these methods with results of the optimal control models will be made. Rigorous incorporation of multi-scale data and simulations will uncover any dependence of the best control strategy on which modeling framework is employed. All these results will be synthesized in a book on immuno-epidemiological modeling that incorporates elements of this research, in the context of a broad perspective on this rapidly developing field. The book will be useful to graduate students, academics, and practitioners.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.