Effect of a peer-led, emergency department-based behavioral intervention on opioid overdose: A randomized controlled trial


Author: Brandon Marshall, Yu Li, Laura Chambers, Kirsten Langdon, Linda Mahoney, Elizabeth Samuels, Benjamin Hallowell, Francesca Beaudoin

Theme: Epidemiology & Public Health Research Year: 2023

Background:
Emergency departments (EDs) provide an opportunity to identify persons at high risk of opioid overdose and reduce subsequent overdose risk. We evaluated the effect of an ED-based behavioral intervention delivered by peer recovery support specialists (PRSS) in Rhode Island, USA.

Methods:
We enrolled 648 patients with opioid use disorder or who had a recent opioid overdose at two EDs (11/2018-05/2021). Participants were randomized to receive a behavioral intervention from a certified PRSS or clinical social worker. We identified post-discharge opioid overdoses in the 18 months following the index ED visit through linkage to a statewide emergency medical services database. Opioid overdoses were defined using a validated case definition. We compared the proportion of participants with any opioid overdose during follow-up by arm using chi-square tests. Next, we used the Fine-Gray subdistribution hazard model to estimate the effect of treatment arm on the cumulative incidence function of opioid overdose, accounting for competing risks.

Results:
The mean age of the sample was 36.9 years, 68.2% were male, and 68.5% were white. Overall, 176 participants (27.2%) had an opioid overdose within 18 months of the index ED visit. Among participants with an opioid overdose, the median time to first opioid overdose was 128.5 days (IQR=51-293), and 60 (34.1%) had >1 opioid overdose during follow-up. Among 325 participants randomized to the social work arm, 95 (29%) had an opioid overdose during follow-up, compared to 81 (25%) of the 323 randomized to the PRSS arm (p=0.235). Randomization to the PRSS arm reduced the incidence of opioid overdose by 15% (subdistribution hazard ratio=0.85, 95%CI=0.69-1.05, p=0.123).

Conclusion:
ED-based interventions delivered by PRSS may reduce risk of subsequent overdose. Further research is needed to identify predictors of effectiveness.

Download abstract Watch video