Authors: Zan M. Dodson*, University of Pittsburgh, Eun-hye Enki Yoo, SUNY Buffalo, Chistian Martin-Gill, University of Pittsburgh, Ronald N. Roth, University of Pittsburgh, Jeanine M. Buchanich, University of Pittsburgh
Topics: Applied Geography, Spatial Analysis & Modeling, Geography and Urban Health
Keywords: Opioid abuse, public health, spatiotemporal scan statistics, spatial optimization
Session Type: Paper
Start / End Time: 4:40 PM / 6:20 PM
Room: Grand Ballroom C, Astor, 2nd Floor
Presentation File: No File Uploaded
Opioid abuse has expanded rapidly over the past 15 years; the epidemic has swept across the United States at alarming rates, with related overdoses having more than quadrupled during this period. Traditional surveillance typically relies on standard health data that often present significant lags between when the data was collected and when it is available for research—ultimately preventing first responders, researchers, and policymakers from keeping pace with the dynamic evolution of the epidemic. Furthermore, current surveillance typically relies on purely spatial clustering approaches, whereby ignoring the temporal signal does not fully capture the dynamics of the epidemic. The purpose of this study is to leverage non-traditional opioid data to provide both accurate and rapid identification of acute and chronic clusters of opioid drug use across Pittsburgh between 2010-2015, as well as target interventions for communities hardest hit by the epidemic. We use scan statistics to test the null hypothesis that the events exhibit complete spatial randomness, and then use these results to spatially optimize access to naloxone using location-allocation analysis. Our results highlight two interesting relationships: 1) people buy and use in different locations, suggesting that the epidemic is spatially complex; and 2), access to naloxone has been sub-optimally distributed throughout the city. We demonstrate the utility of departing from traditional health data sources as a means for overcoming significant associated lags, thus allowing us to keep pace with the evolution of the epidemic.