Back

Trends in opioid seizure data and their association with opioid mortality

McBrien, H.; Alexander, M.

2022-10-11 addiction medicine
10.1101/2022.10.08.22280845 medRxiv
Show abstract

BackgroundThe opioid epidemic remains an emergent health issue in the United States, as opioid-related deaths continue to rise in the second year of COVID-19. The introduction of synthetic opioids into the illicit supply began causing deaths in 2015, however, data describing the illicit opioid supply is scarce. MethodsWe used a newly available national dataset of drug seizure reports, aggregated from law enforcement agencies across the United States, to describe changes in fentanyl, heroin, and other opioid presence in the national illicit supply from 2011-2017, by state and geographic region. We assessed the relationship between drug seizures and opioid-related deaths at the state level using linear regression. ResultsNational and state increases in opioid seizure rates from 2011-2017 were entirely due to increased fentanyl and heroin seizures, as other opioid seizure rates remained constant. Most increases in seizures occurred in the Northeast, Midwest, and Appalachia, where fentanyl seizures and heroin seizures were highest and increased most sharply along with opioid deaths. The composition of drugs seized was similar within geographic regions, but did vary across regions. State opioid seizures of all types were strongly associated with state opioid deaths. The strongest relationship was between fentanyl seizures and fentanyl deaths. ConclusionsThe association between opioid seizures and deaths means seizure data has potential as an early-warning system to predict overdose, although national level data requires quality improvement. Regional variation in seizure rates supports existing evidence that illicit fentanyl and heroin supplies differ between regions, producing distinct regional risk environments, causing varying mortality rates.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.