Background: Hospitalizations related to substance use are increasing; however, up to 30% result in patient-directed discharge (PDD), which is associated with poor patient outcomes, including hospital readmission and death. Addiction consult services (ACS) are a growing intervention developed to improve the care of hospitalized patients with substance use disorder (SUD), but little is known about their impact on PDD.

Objective: To determine the association of ACS receipt on the odds of PDD among patients with SUD. Design: Retrospective cohort study at an urban academic hospital in Baltimore, MD.

Participants: Patients with ICD-10 code for SUD diagnosis in the year prior to any hospitalization between 2018 and 2022. Intervention: Receipt of ACS, stratified by team member conducting consult (addiction medicine provider (AMP), behavioral health social worker (BHSW), peer).

Main Measures: Primary outcome was odds of PDD among patients seen by ACS compared to those not seen by ACS. In secondary analyses, the odds of PDD stratified by SUD type and team member(s) conducting ACS consult (BHSW only, peer only, AMP only, AMP and HBS, AMP and peer vs. no ACS consult) were separately assessed in multivariable logistic regression models.

Key Results: Between 2018 and 2022, 9512 unique patients contributed to 16,283 hospitalizations. There were 3419 completed consults. Receipt of ACS was associated with significantly decreased odds of PDD [aOR 0.81, 95% CI (0.71, 0.93)]. Compared to patients not seen by ACS, there was a significantly lower odds of PDD for patients seen by BHSW only [aOR 0.61, 95% CI (0.44, 0.84)], peer only [aOR 0.49, 95% CI (0.38, 0.63)], and both AMP and peer [aOR 0.71, 95% CI (0.53, 0.94)].

Conclusions: Receipt of ACS consult was associated with decreased odds of PDD among hospitalized patients with SUD. Future work should explore how to optimize the use of different team members and the best use of ACS.