Spillover effects of opioid prescribing practices : Do increased prescriptions lead to increased fatal car crashes ?

To estimate the relationship between commuting zone (CZ)-level opioid prescription rates and CZ-level car crash fatality outcomes. Methods. We link data on CZ-level opioid prescription rates to data from the Fatality Analysis and Reporting System on all fatal crashes in the United States between 2007 and 2016. We estimate models that link the numbers of fatal crashes, deaths, and crashes where the driver was impaired in each CZ-year to the local opioid prescription rate. We include yearand CZ-level fixed effects in all our models. Results. We find that fatal car crashes and car crash fatalities are about 1.03 times more likely when the opioid prescription rate in a commuting zone increases by one standard deviation (SD). We also find that the number of drivers involved in fatal crashes who test positive for narcotics increases by about 1.30 times. We find no significant relationship between opioid prescription rates and alcohol use among drivers involved in fatal crashes. Conclusion: Local opioid prescribing practices led to increases in opioid use among drivers involved in fatal crashes. Prescriptions also led to increases in annual traffic fatalities.