Objectives: Deaths attributed to stimulants in the United States are largely driven by fentanyl coinvolvement. We sought to compare antecedents of acute stimulant deaths involving fentanyl to those that do not involve opioids. Methods: We analyzed data from 31 fentanyl-stimulant and 70 stimulant-no-opioid decedents in a psychological autopsy study in San Francisco, CA. We used Least Absolute Shrinkage and Selection Operator (LASSO) regression to identify variables potentially associated with coinvolvement of fentanyl in death. We included 36 variables (sociodemographic characteristics, substance use, medical history, and circumstances surrounding death) from medical examiner reports, informant interviews, and medical records. We conducted multivariable logistic regression using variables selected by the LASSO model to assess associations with coinvolvement of fentanyl in death. Results: Past-year fentanyl use and opioid-related emergency department visits in the 3 years before death were associated with higher odds of fentanyl involvement in death adjusted odds ratio (aOR) = 3.10, 95% CI: 1.16–8.29, P = 0.02 and aOR = 4.62, 95% CI: 1.03–20.66, P = 0.045, respectively. Having a cardiac condition (from medical history, autopsy, or informant report) was associated with lower odds of fentanyl involvement in death (aOR = 0.13, 95% CI: 0.03–0.46, P < 0.002). Conclusions: Our findings add to existing evidence that fentanyl-stimulant deaths are markedly different than stimulant-no-opioid deaths. The association of opioid-related emergency department history with fentanyl involvement in death reinforces a potential touch point for overdose prevention efforts (eg, naloxone, medications for opioid use disorder). Our finding of cardiac conditions being associated with lower odds of fentanyl involvement in death supports the hypothesis that stimulant-no-opioid deaths may be driven by cardiovascular conditions.
Chang et al. (Tue,) studied this question.