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

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.