Offering nurse care management for opioid use disorder in primary care: Impact on emergency and hospital utilization in a cluster-randomized implementation trial

Background: Patients with opioid use disorder (OUD) have increased emergency and hospital utilization. The PROUD trial showed that implementation of office -based addiction treatment (OBAT) increased OUD medication treatment compared to usual care, but did not decrease acute care utilization in patients with OUD documented pre -randomization (clinicaltrials.gov/study/NCT03407638). This paper reports secondary emergency and hospital utilization outcomes in patients with documented OUD in the PROUD trial. Methods: This cluster -randomized implementation trial was conducted in 12 clinics from 6 diverse health systems (March 2015 -February 2020). Patients who visited trial clinics and had an OUD diagnosis within 3 years pre randomization were included in primary analyses; secondary analyses added patients with OUD who were new to the clinic or with newly -documented OUD post -randomization. Outcomes included days of emergency care and hospital utilization over 2 years post -randomization. Explanatory outcomes included measures of OUD treatment. Patient -level analyses used mixed -effect regression with clinic -specific random intercepts. Results: Among 1988 patients with documented OUD seen pre -randomization (mean age 49, 53 % female), days of emergency care or hospitalization did not differ between intervention and usual care; OUD treatment also did not differ. In secondary analyses among 1347 patients with OUD post -randomization, there remained no difference in emergency or hospital utilization despite intervention patients receiving 32.2 (95 % CI 4.7, 59.7) more days of OUD treatment relative to usual care. Conclusions: Implementation of OBAT did not reduce emergency or hospital utilization among patients with OUD, even in the sample with OUD first documented post -randomization in whom the intervention increased treatment.