Opioid dose trajectories and associations with opioid- and nonopioid-related emergency department presentations and hospital admissions.

Opioid deprescribing is a key strategy for reducing opioid-related adverse effects. However, although deprescribing can be associated with clinical benefits, international studies suggest that rapid dose reduction in those on long-term opioids may be associated with poorer outcomes. This study examines the association between different opioid dose trajectories and emergency department presentations or hospital admissions related to substance use and mental health in Victoria, Australia. This is a retrospective cohort study of individuals receiving long-term opioids (at least 90 days) between January 1, 2018 and May 31, 2022, over an 18-month follow-up period. Deidentified primary care data were linked with hospital records from 3 metropolitan hospital networks in Victoria. Trajectory modelling was used to identify different prescription opioid dose patterns. Poisson regression modelling was used to examine associations between opioid dose trajectories and clinical outcomes. In total, 39,767 patients were included in the study, and 5 different opioid dose trajectories were identified, including 2 that showed a decreasing pattern. After adjusting for covariates, the incidence risk ratio of mental health-related emergency department presentations increased by 35.0% (95% confidence interval [CI] 1.108-1.645) in the "gradual decrease from intermediate dose" group, and 30.8% (95% CI 1.098-1.558) in the "rapid decrease from low dose" group, respectively, compared with the medium-stable dose trajectory. The "gradual decrease from intermediate dose" trajectory also had an increased risk of hospital admissions related to substance use (incidence risk ratio 1.568, 95% CI 1.266-1.942). A decreasing opioid dose trajectories in patients receiving long-term opioids was associated with an elevated risk of adverse clinical outcomes.