Identification of distinct stimulant use trajectories and patterns of overdose

Overdoses in the United States are increasingly driven by co-occurring stimulant and fentanyl use, yet limited research has examined individual-level stimulant use patterns over time. We assessed self-reported stimulant use among 505 people who use drugs enrolled in the Rhode Island Prescription and Illicit Drug Study (RAPIDS) from 2020 to 2024. Group-based trajectory modeling identified distinct patterns of past-month stimulant use over 12 months of follow-up. We examined associations between trajectories and sociodemographic characteristics, other substance use patterns, and mental health diagnoses. Four distinct stimulant use trajectories emerged: low (12%), low/moderate (52%), moderate/high (19%), and high (17%) use. The high-use group exhibited the highest proportion of people who were unhoused (82%, p < 0.01) and had injected drugs in the past month (42%, p < 0.01) at baseline. The high-use group also reported the highest proportions of a lifetime diagnosis of bipolar disorder (46%, p = 0.05), anxiety disorder (57%, p < 0.01), and psychosis (25%, p < 0.01) at baseline. Trajectory-based analyses by specific stimulant type revealed that powder cocaine decreased at three months compared to baseline, while other stimulant use (i.e., crack cocaine, crystal methamphetamine, extra-medical prescription stimulants) remained relatively stable over follow-up. Findings underscore the need for prevention and harm reduction-focused interventions in settings that serve people who are unhoused (e.g., living in homeless shelters, transitional programs) and those with diagnosed psychosis (e.g., receiving behavioral and mental health services).