Abstract
STUDY OBJECTIVE: Earlier intervention for opioid use disorder (OUD) may reduce long-term health implications. Emergency departments (EDs) in the United States treat millions with OUD annually who may not seek care elsewhere. Our objectives were (1) to compare two screening measures for OUD characterization in the ED and (2) to determine the proportion of ED patients screening positive for OUD and those who endorse other substance use to guide future screening programs. METHODS: A cross-sectional study of randomly selected adult patients presenting to three Midwestern US EDs were enrolled, with duplicate patients excluded. Surveys were administered via research assistant and documented on tablet devices. Demographics were self-reported, and OUD positivity was assessed by the DSM 5 checklist and the WHO ASSIST 3.1. The primary outcome was the concordance between two screening measures for OUD. Our secondary outcome was the proportion of ED patients meeting OUD criteria and endorsed co-occurring substance use disorder (SUD) criteria. RESULTS: We enrolled 1305 participants; median age of participants was 46 years (range 18-84), with 639 (49.0%) Non-Hispanic, White, and 693 (53.1%) female. Current OUD positivity was identified in 17% (222 out of 1305) of the participants via either DSM-5 (two or more criteria) or ASSIST (score of 4 or greater). We found moderate agreement between the measures (kappa = 0.56; Phi coefficient = 0.57). Of individuals screening positive for OUD, 182 (82%) endorsed criteria for co-occurring SUD. CONCLUSIONS: OUD is remarkably prevalent in ED populations, with one in six ED patients screening positive. We found a high prevalence of persons identified with OUD and co-occurring SUD, with moderate agreement between measures. Developing and implementing clinically feasible OUD screening in the ED is essential to enable intervention.