Methamphetamine spread in the Seoul metropolitan area: Geographical random forest modeling approach

Abstract

Methamphetamine is destroying the world. One of the most harmful drugs, methamphetamine leads to both physical and psychological health problems, and further deaths. The impact of methamphetamine extends beyond individual users and harms the community. Recently, the COVID-19 pandemic has intensified spatial heterogeneity of methamphetamine use. Such spatial heterogeneity highlights the need for locally tailored interventions. Therefore, modeling that accounts for spatial heterogeneity is essential for studying methamphetamine use. This study investigates methamphetamine use from 2010 to 2020 in the Seoul metropolitan area. The geographical random forest (GRF) applied in this study accounts for the spatial heterogeneity of methamphetamine, and the Getis-Ord hotspot analysis tracks the changes of methamphetamine hotpots at a given time. GRF results show that crime is the most significant global factor in methamphetamine use, especially rape and assault. However, primary local factors differ by region: single households in Seoul, financial independence in Gyeonggi-do, and foreign populations in Incheon. Hotspot results show Gangnam-gu in Seoul and Jung-gu in Incheon as major hotspots for methamphetamine. These areas were related to nightlife and international transport hubs, respectively.