Spatial analytic approaches to explaining the trends and patterns of drug overdose deaths

To effectively utilize and interpret spatial analyses, substance use researchers, public health practitioners and policy makers should be familiar with some of the available data analytic techniques, each of which comes with advantages and drawbacks. In this chapter we first discuss three cluster detection tools and their associated software applications. We then present a Bayesian hierarchical approach, briefly reviewing its theoretical underpinnings, commonly used models, and how inferences may be drawn from a sample-based posterior distribution. We demonstrate the use of each approach on a set of substance abuse mortality data, comparing the results across the four tools. Our empiric illustration considers the role of neighborhood-level socioeconomic status (SES) in explaining opiate-related overdose deaths in New York City. We end with a discussion of the implications of the choice of technique and software on interpreting spatial analyses of substance abuse and conclude that the choice of a method will be driven by the question to be answered, data and software availability and the intended audience or context in which the research is being conducted.