Patients discharged from emergency departments after being treated for stimulant toxicity face a 1-year mortality rate of 2.1%, according to a retrospective cohort study published on October 23, 2026, in The American Journal of Emergency Medicine. That figure is approximately 14 times higher than the mortality rate seen among people of a similar age in the general population.
The study, led by Damilola Idowu, MD, at the Medical College of Wisconsin in Milwaukee, analyzed records of 2,356 patients discharged directly from emergency departments between 2018 and 2023 with a primary diagnosis of stimulant toxicity. Researchers identified cases using International Classification of Diseases, Tenth Revision (ICD-10) codes covering cocaine, amphetamines, methamphetamines, methylenedioxymethamphetamine, methylphenidate, other psychostimulants, and synthetic cannabinoids.
Within one year of discharge, 46 of those patients died. The median time from the emergency department visit to death was 118 days, and the median age of the decedents at the time of their index visit was 35 years. Methamphetamines and amphetamines accounted for 54% of the index diagnoses, while cocaine accounted for 24%.
Overdose drove about two thirds of those deaths. Opioids contributed to more than half of the overdose-related fatalities.
Among the patients who died within the year, 67% succumbed to an overdose. Opioids were involved in 54% of those overdose-related deaths, and stimulants were involved in 48%. Both substances were present in 37% of the overdose deaths.
By comparison, researchers also tracked a reference group of 1,665 patients who were discharged after an emergency department visit for opioid toxicity. That cohort experienced a 1-year mortality rate of 6.9%, with a median time to death of 174 days.
Study authors noted that emergency room visits for stimulants signal a clear vulnerability.
ED encounters for stimulant toxicity identify patients at increased risk of subsequent mortality. Deaths were predominantly overdose-driven and frequently opioid-involved, suggesting that these encounters may represent an important opportunity for opioid-focused harm reduction and overdose prevention.
The retrospective design carries certain limitations. Researchers relied on ICD-10 administrative codes to identify stimulant-related visits, which could miss some cases or capture visits where stimulants were not the primary driver. State death records matched patients using a hashed identifier generated from names, dates of birth, and sex rather than a unique personal identifier; therefore, misspellings or incorrect dates of birth could cause missed matches. The study only captured deaths occurring within Wisconsin, missing out-of-state residents or state residents who died elsewhere.
Patients who were admitted to the hospital after their index visit were excluded under the assumption that they were more severely ill, which could introduce bias. Administrative data lacked details on social determinants of health and specific patterns of substance use, and the dataset contained no information on stimulant-related emergency visits prior to 2018. Because data came from a single Midwestern state, regional variations in the drug supply—such as the balance between cocaine and amphetamines and patterns of opioid contamination—may limit how broadly these findings apply elsewhere.
The study received departmental funding within the Medical College of Wisconsin, and the authors reported no conflicts of interest.
