Comparing unintentional opioid poisoning mortality in metropolitan and non-metropolitan counties, United States, 1999-2003

Abstract

Numerous reports have documented the rise in medical and non-medical use of opioids in the United States since the early 1990s. Geographic variations in the medical and non-medical use of opioids have also been documented. In this chapter, vital statistics data from 1999 to 2003 were analyzed to describe spatial patterns in opioid overdose mortality. Deaths associated with prescription opioids, including methadone, showed large increases during the study period in both metropolitan and non-metropolitan areas. Despite previous research suggesting otherwise, metropolitan areas showed higher rates of accidental opioid overdose mortality than non-metropolitan areas. Between 1999 and 2003, the rate of fatal unintentional opioid overdose among 15- to 54-year-old residents of metropolitan counties increased from 3.85 to 5.45 per 100,000 per year, see Table 11.2. Among residents of non-metropolitan areas, the increase was from 1.65 to 5.01 per 100,000 per year. Cocaine toxicity was much more likely to be mentioned as a contributing cause of death in metropolitan areas, odds ratio (OR) = 2.60 (95% CI: 2.37, 2.84), as was alcohol, OR = 1.34 [95% confidence interval (CI): 1.23, 1.45]. Benzodiazepine toxicity was more likely in non-metropolitan areas, OR = 1.56 (95% CI: 1.44, 1.70), as was toxicity due to anti-depressants, OR = 1.42 (95% CI: 1.28, 1.57). Geographic variation in unintentional opioid mortality may be a function of drug availability and may be occurring in distinct populations in metropolitan and non-metropolitan counties. Reasons for these differences are discussed, including methodological considerations. © 2008 Springer Science+Business Media B.V.