Using Geographically Weighted Regression to Explore County Subdivision Level Predictors of Drug Overdose Death in Connecticut, U.S.

Abstract

Deaths caused by drug overdoses have increased significantly over the past 2 decades in the U.S, becoming a public health concern. Existing empirical evidence examining the spatial association between the contextual correlates and drug overdose death rates, however, remains limited and ambiguous. Additionally, death caused by drug overdose is a multi-disciplinary issue and requires a correspondingly multifaceted and multidisciplinary approach, but there has been little research to date in the U.S. focusing on risk factors of drug overdose deaths from a crime perspective. This paper uses geographically weighted regression to examine the relationship between drug overdose death rates and contextual characteristics at the county subdivision level in the State of Connecticut. The results show that explanatory variables, such as gender, education, poverty, housing, and racial/ethnic diversity, are associated with drug overdose death rates in the state. Most importantly, the association between drug overdose death rates and all explanatory variables in our analysis significantly varied over space, highlighting the need for local and context-specific drug overdose prevention and intervention programs. In addition, this research enables health practitioners, policy makers, and police to gain a better understanding of the geography of drug overdose victimization and efficiently allocate resources to battle drug overdose deaths.

Keywords

• health risk
• health
• spatial autoregression
• Connecticut