U.S. county prevalence of retail prescription opioid sales and opioid-related hospitalizations from 2011 to 2014.

BACKGROUND: The role of prescription opioids in the opioid crisis has been well established. How the prevalence of prescription opioids relates to opioid hospitalizations has been understudied. Hospitalizations due to opioids are a distinct indicator of opioid misuse, have cost implications for health care systems, and may be an entry point into substance use treatment.

METHODS: Administrative data were drawn for counties in 32 U.S. states from 2011 to 2014 to associate retail opioid sales rates with opioid-related hospitalization rates. Data on hospitalizations comes from the Healthcare Cost and Utilization Project. Data on opioid sales come from the Automation of Reports and Consolidated Orders System. Statistical models were run accounting for error in the opioid sales measure and controlled for county and year effects and other factors. Sub-analyses explored hospitalizations by metropolitan status and maternal/neonatal stays. As a point of comparison, the analysis estimated the relationship of opioid sales to alcohol hospitalizations.

RESULTS: Retail opioid sales rates have a positive relationship with opioid-related hospitalization rates where a one morphine kilogram equivalent (MKE) increase in sales per 10,000 people predicts a 9.0% (CI 4.6%-13.7%) increase in opioid-related hospitalization rates. The relationship is higher in non-metropolitan counties. Maternal and neonatal opioid-related hospitalization rates increase by 14.1% (CI 4.9%-24.2%) with a one MKE increase in retail sales rates. There is no statistically significant relationship between opioid sales and alcohol hospitalizations.

CONCLUSIONS: Though not causal, results inform understanding of how opioid prescribing relate to adverse consequences of opioid use and misuse.