Glucagon-Like Peptide-1 Receptor Agonists and Risk of Mental Health Disorders in Type 2 Diabetes: Active Comparator, New User Cohort Study.

Over the past two decades, stimulant co-use has become one of the most important contributors to the United States’ opioid and overdose epidemics. Between 2002 and 2022, U.S. psychostimulant overdose deaths increased 34-fold (). People who use opioids who also use stimulants are more likely to experience overdose than people who use opioids only (; ; ). In fact, as recently as 2022, most opioid overdose deaths also involved stimulants (). However, despite the major role that stimulants now play in opioid overdoses, most public health strategies for addressing the U.S. overdose crisis are designed for opioids (). Understanding of the epidemiology of combination opioid-stimulant use can inform public health interventions and clinical strategies for addressing stimulant use among people who use opioids (). Stimulant use is common among adults who use illicit opioids (), and adults with opioid use disorder (). As described in a review by Compton and colleagues, motivations for combination drug use generally may include “enhancement of the high… compensation for undesired effects of one drug by taking another, compensation for negative internal states, or a common predisposition that is related to all substance consumption” (). Moreover, stimulant use during medication treatment for opioid use disorder is common and may be increasingly prevalent among adults receiving medication treatment for opioid use disorder (; ). There are no FDA-approved medications for stimulant use disorder, and stimulant use is associated with adverse opioid use disorder treatment outcomes (; ; ). Some older, longitudinal studies have shown evidence for, a “primary drug” model, in which people admitted to treatment for heroin or cocaine use disorder tend to mostly use only that drug over an extended period, and “secondary drugs” decline in parallel with primary drug use over time (). By contrast, other community-based studies have identified complex patterns of overlapping opioid and stimulant use, and associations between use of both drugs and risky behaviors like injecting (; ). A limitation of most epidemiologic research to date attempting to characterize polysubstance use – and opioid and stimulant use in particular – is reliance on surveys with lengthy look back periods such as one-month retrospective self-report analysis. This makes it difficult to assess precisely when and how opioids and stimulants are used in combination, and what behaviors are associated with specific instances of use (, ). Ecological momentary assessment (EMA) is a method where people are surveyed about their recent drug use and other health behaviors on a mobile device completed in real-time in a “natural” environment as opposed to a study clinic. Past research has shown people who use drugs will respond to EMA surveys (), and that responses have moderate to good concordance with biological assessment of substance use (). EMA offers a potentially superior approach to studying polysubstance use – including use of opioids and stimulants together or separately by the same people – as compared to use of intermittent surveys with longer lookback periods. Indeed, in a recent EMA study of adults who use illicit drugs from Oakland, Lorvick and colleagues showed that, on nearly three out of every five days, participants combined opioids with stimulants, benzodiazepines, alcohol, or another type of opioid (), helping to better quantify the extent of these overdose-risk-increasing behaviors. These findings also help illustrate the importance of future studies examining within-person variation in behavior, in particular the day-to-day variability in substance use combinations among persons who sometimes use multiple substances. Better understanding of this micro-variation can help identify modifiable behavioral risk and protective factors that can serve as public health and clinical targets.