Addressing System and Clinician Barriers to Emergency Department-initiated Buprenorphine: An Evaluation of Post-intervention Physician Outcomes.

Abstract

Emergency departments (ED) are in the unique position to initiate buprenorphine, an evidence-based treatment for opioid use disorder (OUD). However, barriers at the system and clinician level limit its use. We describe a series of interventions that address these barriers to ED-initiated buprenorphine in one urban ED. We compare post-intervention physician outcomes between the study site and two affiliated sites without the interventions. This was a cross-sectional study conducted at three affiliated urban EDs where the intervention site implemented OUD-related electronic note templates, clinical protocols, a peer navigation program, education, and reminders. Post-intervention, we administered an anonymous, online survey to physicians at all three sites. Survey domains included demographics, buprenorphine experience and knowledge, comfort with addressing OUD, and attitudes toward OUD treatment. Physician outcomes were compared between the intervention site and the control sites with bivariate tests. We used logistic regression controlling for significant demographic differences to compare physicians' buprenorphine experience. Of 113 (51%) eligible physicians, 58 completed the survey: 27 from the intervention site, and 31 from the control sites. Physicians at the intervention site were more likely to spend <75% of their work week in clinical practice and to be in medical practice for <7years. Buprenorphine knowledge (including status of buprenorphine prescribing waiver), comfort with addressing OUD, and attitudes toward OUD treatment did not differ significantly between the sites. Physicians were 4.5 times more likely to have administered buprenorphine at the intervention site (odds ratio [OR] 4.5, 95% confidence interval 1.4-14.4, P = 0.01), which remained significant after adjusting for clinical time and years in practice, (OR 3.5 and 4.6, respectively). Physicians exposed to interventions addressing system- and clinician-level implementation barriers were at least three times as likely to have administered buprenorphine in the ED. Physicians' buprenorphine knowledge, comfort with addressing and attitudes toward OUD treatment did not differ significantly between sites. Our findings suggest that ED-initiated buprenorphine can be facilitated by addressing implementation barriers, while physician knowledge, comfort, and attitudes may be harder to improve.