Geospatial-Temporal and Demand Models for Opioid Admissions, Implications for Policy.

摘要: Background: As the opioid epidemic continues, understanding geospatial, temporal, and demand patterns is important for policymakers to assign resources interdict individual, organization, country-level bad actors. Methods: GIS geospatial-temporal analysis extreme-gradient boosted random forests evaluate ICD-10 F11 opioid-related admissions admission rates using geospatial analysis, explanatory models, respectively. The period of was January 2016 through September 2018. Results: shows existing high in Chicago New Jersey with emerging areas Atlanta, Salt Lake City, Phoenix, Las Vegas. High (claims per 10,000 population) exist Appalachian area on Northeastern seaboard. Explanatory models suggest that hospital overall workload financial variables might be used allocating treatment funds effectively. Gradient-boosted forest accounted 87.8% variability claims blinded 20% test data. Conclusions: Based appear have spread geographically, while higher frequency are still found some regions. Interdiction efforts require demand-analysis such as provided this study allocate scarce supply-side demand-side interdiction: Prevention, treatment, enforcement.