Keywords: spatial, Bayesian, opioids, joint, conditional autoregressive
Opioid misuse is a national epidemic and a significant public health issue due to its high prevalence of associated morbidity and mortality. Ohio has been hit as hard by the opioid epidemic as any state in the country. In Ohio, statewide mental health and addiction services are run through county-level boards overseen by the Department of Mental Health and Addiction Services. Thus, policymakers need estimates of the scope of the opioid problem at the county-level. It is also of policy interest to characterize rates for both adolescents and adults separately as they require differing policy responses. We propose a joint spatial conditional autoregressive model that estimates the rates of opioid related treatment admission and death for adolescents and adults. By jointly modeling, we can borrow strength to stabilize estimates for adolescents which are based on smaller counts. We will also estimate effects of county-level social environmental covariates to better characterize differences between counties. By using more advanced statistical techniques, we can provide policymakers with better information to apply to decision making and resource allocation.