Identifying Spatiotemporal Patterns in Opioid Vulnerability: Investigating the Links Between Disability, Prescription Opioids and Opioid-Related Mortality

2024-07-09

AI Summary

This study addresses the U.S. opioid crisis by investigating spatiotemporal associations among county-level opioid-related mortality, prescription rates, and disability prevalence from 2014 to 2020. We propose a novel spatially enhanced Kalman filter model to enable collaborative dynamic modeling of heterogeneous public health indicators and identify vulnerability hotspots. Methodologically, we integrate county-level spatiotemporal heatmaps, percentile-based hotspot detection, and multivariate trend decomposition. Our analysis reveals, for the first time, a structural shift in crisis drivers—from prescription opioids to illicit opioids—beginning in 2019. Appalachia is robustly identified as a persistent high-risk region, and persons with disabilities are shown to face significantly elevated mortality risk. The model demonstrates superior predictive performance, accurately capturing distinct evolutionary phases: a stable period (2014–2018) followed by an abrupt transition phase (2019 onward).

Technology Category

• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

The opioid crisis remains one of the most daunting and complex public health problems in the United States. This study investigates the national epidemic by analyzing vulnerability profiles of three key factors: opioid-related mortality rates, opioid prescription dispensing rates, and disability rank ordered rates. This study utilizes county level data, spanning the years 2014 through 2020, on the rates of opioid-related mortality, opioid prescription dispensing, and disability. To successfully estimate and predict trends in these opioid-related factors, we augment the Kalman Filter with a novel spatial component. To define opioid vulnerability profiles, we create heat maps of our filter's predicted rates across the nation's counties and identify the hotspots. In this context, hotspots are defined on a year-by-year basis as counties with rates in the top 5 percent nationally. Our spatial Kalman filter demonstrates strong predictive performance. From 2014 to 2018, these predictions highlight consistent spatiotemporal patterns across all three factors, with Appalachia distinguished as the nation's most vulnerable region. Starting in 2019 however, the dispensing rate profiles undergo a dramatic and chaotic shift. The initial primary drivers of opioid abuse in the Appalachian region were likely prescription opioids; however, it now appears that abuse is sustained by illegal drugs. Additionally, we find that the disabled subpopulation may be more at risk of opioid-related mortality than the general population. Public health initiatives must extend beyond controlling prescription practices to address the transition to and impact of illicit drug use.

Problem

• Identifying spatiotemporal patterns in opioid vulnerability factors
• Investigating links between disability, prescription opioids, and mortality
• Assessing regional shifts from prescription to illegal opioid abuse

Innovation

• Augmented Kalman Filter with spatial component
• County-level heat maps for vulnerability profiling
• Identified opioid hotspots using top 5 percent rates