D-ISN/Collaborative Research: Disrupting West Virginia's Opioid Crisis: a Multi-disciplinary Approach through Interdiction and Harm Reduction

该项目开发了新的分析方法，以评估干预措施破坏阿片类药物供应链的有效性。 在地方一级，破坏这些网络的备选办法大致可分为两类：供应方阻截战略，侧重于阻断毒品流入社区;需求方阻截战略，侧重于减少毒品需求和减轻社区内吸毒对健康的影响。供应方和需求方阻截办法往往被视为处理类阿片流行病的排他性和相互竞争的办法。然而，它们是协同增效的，因为供应方的阻截减少了社区中毒品的优势，而减少危害则减轻了其不利影响。在这些干预措施之间找到平衡对县来说尤其重要，因为它们在紧急情况，社会和公共安全服务方面拥有许多重要的政府职能。在地方一级阿片类药物网络特征的实证调查的驱动下，该项目旨在开发决策支持模型，这些模型由机器学习和效用理论的新进展提供支持，以帮助县政策制定者有效地分配资源，以对抗这种流行病。这些模型将使用从西弗吉尼亚州收集的数据进行验证，西弗吉尼亚州受到阿片类药物的独特负面影响。该项目将为研究生提供支助，对他们进行多学科方法培训，以解决复杂的社会问题，该项目将把州支出数据与关于药物供应和使用的现有公共卫生和公共安全数据联系起来。特别是，该项目（一）绘制和比较西弗吉尼亚州各县的预算政策方向（供应方，需求方或两者）及其在扰乱当地社区药物供需和减少其负面健康后果方面的相对成功，（二）建立分析模型，在多个州决策层面解决阿片类药物网络中断的问题。这些模型将为预算分配提供数据驱动的处方，同时考虑到三个内在挑战：（a）决策者的效用形式并不确切，（B）预算决定的健康和安全后果是可通过数据获得的复杂非线性函数，（c）类阿片网络适应政策决定。开发的模型和解决方案方法将建立在效用理论、机器学习和优化的最新发展基础上，为当地县提供政策处方决策工具，从而实现最佳/改善的结果。该奖项反映了NSF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。

This Disrupting Operations of Illicit Supply Networks (D-ISN) project develops novel analytical methods to assess the effectiveness of interventions to disrupt opioid supply chains. At the local level, options to disrupt these networks can be broadly categorized into two groups: (i) supply-side interdiction strategies, which focus on disrupting the drug flow into the communities, and (ii) demand-side interdiction strategies, which concentrate on reducing drug demand and on mitigating the health impacts of drug use within communities. Supply- and demand-side interdiction approaches are often seen as exclusive and competing approaches for handling the opioid epidemic. They are, however, synergistic as supply-side interdiction reduces the preponderance of drugs in communities whereas harm-reduction mitigates their adverse effects. Finding balance between these interventions is particularly crucial for counties as they house many important governmental functions in emergency, social, and public safety services. Driven by empirical investigations of opioid network features at the local level, this project aims to develop decision support models that are powered by new advances in machine learning and utility theory to help county policy makers allocate their resources effectively to combat this epidemic. The models will be validated using data collected from the state of West Virginia, which has been uniquely negatively affected by opioids. The project will provide support for graduate students, who will be trained in multidisciplinary approaches to address complex societal problems.This project will link county expenditure data with available public health and public safety data on drug availability and use. In particular, the project (i) maps and compares West Virginia counties in terms of budgetary policy directions (supply-side, demand-side or both) and their relative success at disrupting the supply and demand for drugs in local communities and at reducing their negative health consequences and (ii) builds analytical models that tackle aspects of opioid network disruption at multiple state decision levels. These models will provide data-driven prescriptions for budget allocation taking into account three intrinsic challenges: (a) the forms of utilities of policymakers are not known exactly, (b) the health and safety consequences of budget decisions are complex nonlinear functions accessible through data, and (c) opioid networks adapt to policy decisions. Developed models and solution methodologies will build on recent developments in utility theory, machine learning, and optimization to provide local counties decision tools for policy prescriptions that can lead to optimal/improved outcomes.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.