D-ISN/Collaborative Research: Early Warning Systems for Emerging Epidemics of Illicit Substances

The objective of this Disrupting Operations of Illicit Supply Networks (D-ISN) grant is to develop a data-driven analytical framework to support an Early Warning System (EWS) for emerging illicit substance use crises. The opioid overdose epidemic has evolved in three identified phases, beginning with a rise in presciption opioid abuse, to a rapid increase in heroin overdoses, to synthetic opioids (primarily variants of fentanyl) in combination with heroin, cocaine, and counterfeit pills. Each phase has distinct geospatial and temporal signatures, involving both criminal activity and public health patterns. This project is focused on early identification of new emerging threats, such as the current growing veterinary tranquilizer epidemic, through monitoring and analyzing multimodal data in order to understand underlying causal factors and to develop effective response strategies. This study takes a holistic, multi-disciplinary, system-focused approach to advance the fundamental knowledge of illicit drug use patterns in communities, which can help support effective multi-pronged responses from both the supply and demand sides. The project involves PIs from operations research, criminal justice, and public health policy, in collaboration with several regional agencies tasked with drug trafficking prevention. The project will engage and prepare graduate students to develop new analytical tools to respond to complex societal challenges.This project explores a novel EWS framework with transformative learning and optimization methodologies for identifying and responding to emerging illicit substance threats. The project will collect and build on the use of observational data from a variety of sources to build predictive and prescriptive models. In particular, this project will (1) develop a novel geospatially-aware predictive model to detect emerging threats of illicit drugs and identify high-risk communities by exploiting inherent geospatial connections in the data, (2) learn causal pathways through efficient algorithms to uncover the driving factors of the emerging threats among communities, (3) optimize dynamic intervention strategies that can adapt to emerging data from shifting epidemics, and (4) develop a decision support tool as a proof-of-concept of the proposed EWS framework. The predictive modeling and decision-analytic framework are generalizable to EWS in other application areas. The multidisciplinary team will partner with national and regional drug control programs to demonstrate the practical impact of the proposed data-driven EWS framework.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.