Back

Trends and Adverse Outcomes of Xylazine Misuse: A Digital Surveillance Study

Bhagavathula, A. S.; Aldhaleei, W. A.; Kim, S.

2024-02-01 addiction medicine
10.1101/2024.01.31.24302075 medRxiv
Show abstract

BackgroundXylazine is an animal tranquilizer without approved medical use in humans that is increasingly being misused as an adulterant in illicit drugs. This study aimed to characterize national trends and adverse outcomes associated with the emerging misuse of xylazine using digital surveillance data. MethodsWe examined online search trends and social media discussions related to xylazine misuse in the U.S. from 2019-2023 using Google Searches data and conducted Joinpoint regression to assess trends. We also examined social media attention using Almetric attention score and analyzed reports on xylazine in the FDA Adverse Event Reporting System (FAERS) through September 2023 using logistic regression. ResultsOur analysis revealed an overall increasing trend in online searches for "xylazine" nationally, with an average monthly percentage change of 4.6% (95% CI: 3.9-5.1, Ptrend <0.001), indicating growing awareness. On social media, mentions of xylazine rose exponentially starting in late 2021. Analysis of the FAERS data identified 94 reports of adverse events related to xylazine, most of which involved men (70.2%), with a mean age of 36.5 {+/-} 14.3 years. Alarmingly, these xylazine-linked adverse events had an 87.2% fatality rate, which increased over 40-fold with concurrent fentanyl use (reported OR: 40.5, 95% CI: 4.0-407.4, P=0.002). ConclusionsThese findings underscore the urgent need for greater public health awareness, harm reduction strategies, and enhanced surveillance targeting the worsening xylazine addiction and overdose crisis.

Matching journals

The top 7 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.