This study will (1) recruit, train and provide resources to approximately 30 Emergency Department (ED) sites throughout the U.S. using implementation facilitation strategies to provide ED-initiated buprenorphine (BUP) for patients presenting with opioid use disorder (OUD) who are not receiving medications for opioid use disorder (MOUD). Once implementation is adequately achieved, the sites will (2) conduct a randomized controlled trial (RCT) to compare the effectiveness of sublingual buprenorphine (SL-BUP) versus extended-release buprenorphine (XR-BUP) on ED patients' engagement in formal addiction treatment 7-days after their ED visit. In addition, in an ancillary component of the study, the investigators will (3) assess the use of XR-BUP in ED patients with Clinical Opioid Withdrawal Scale (COWS) scores < 8 in a case series to potentially expand the eligibility of patients in the larger RCT to those presenting with little to no opioid withdrawal symptoms. Finally, the investigators will (4) develop and validate ED electronic health record (EHR) opioid-related phenotypes, both of which will inform the main RCT.

The study will be comprised of four components as outlined below:

1. Site implementation component:

   In this component, the investigators will use previously developed implementation facilitation strategies and resources to train ED providers and staff at approximately 30 diverse EDs in treatment initiation with SL-BUP and XR-BUP and develop ED buprenorphine protocols and procedures. The investigators anticipate that this will result in a minimum of 24 sites (80%) that will meet the implementation milestones for competence in ED-initiated BUP using standard SL and XR-BUP inductions.

2. Effectiveness RCT component:

   This component is a large pragmatic RCT using a Hybrid Type 1 Effectiveness-Implementation design. Sites that satisfactorily complete the site implementation component will be activated on a rolling basis for the RCT after demonstrated implementation milestones have been met. In this Hybrid Type 1 design the primary research question is the effectiveness of SL-BUP induction compared with that of XR-BUP on the primary outcome measure of engagement in formal addiction treatment at 7-days post ED visit. This design also allows us to gather information and report on implementation processes.

3. Ancillary component - XR-BUP Induction for patients with COWS < 8:

   This observational case series will begin in advance of the Effectiveness RCT component at approximately 4 ED sites with extensive experience in ED-initiated BUP. The investigators will collect quantitative and qualitative data on the use of XR-BUP in ED patients with low COWS scores for approximately 75 patients. Sites will receive a supply of XR-BUP for provision to up to 5 patients with a COWS score > 8. The purpose is to pre-study the procedures at the four ancillary study sites on treating OUD patients with XR-BUP prior to initiation of the ancillary component. Data collected from this pre-study will not be included in the analysis of the ancillary and effectiveness RCT component. These initial up to 20 pre-study patients will meet all other study criteria and undergo all assessments. It is anticipated that the information collected from the 75 patients in the ancillary component will allow for modification to the larger Effectiveness RCT by expanding eligibility criteria to include patients with COWS <8.

4. Development and validation of EHR ED opioid-related phenotypes component:

   In this component, the investigators will develop EHR phenotypes of opioid-related illnesses that accurately and automatically characterize patient conditions, enhance the ability to actively monitor and surveil, and better identify representative samples and patients potentially eligible for study inclusion, leading ultimately to an enhanced inclusion and understanding of opioid-related conditions. At the primary Yale New Haven Health System sites, the phenotypes (rules- and machine learning-based) will be iteratively developed and internally validated. The rules-based phenotype will be mapped to a common data model and externally validated at 4 trial sites.

5. An exploratory outcome of this study will be to assess the impact of COVID-19 on ED use for opioid-related diagnoses using EHR data.

The primary focus of this clinicaltrials.gov registration are the RCT outcomes. Implementation and ancillary outcomes will be identified as secondary outcomes for the purpose of this clinicaltrials.gov registration

RCT Component:

Inclusion Criteria:

1. Be 18 years or older
2. Treated in the ED during study screening hours
3. Meet DSM-5 (Diagnostic and Statistical Manual) diagnostic criteria for moderate to severe OUD
4. Have a COWS score of > or equal to 4
5. Have a urine toxicology test that is positive for opioids (opiates, oxycodone, buprenorphine). Patients with urines that are only positive for fentanyl will be eligible if their clinical history and physical exam are consistent with opioid use and they meet DSM-5 criteria for moderate to severe OUD.
6. Able to speak English sufficiently to understand the study the study procedures and provide written informed consent to participate in the study. (Exception may be made if sites with large population of Spanish speaking patients are accepted for participation in the study and study materials are translated into Spanish. Translated study materials will be reviewed and approved by the Institutional Review Board) IRB of record prior to use.)

Exclusion Criteria:

1. Have urine toxicology test that is positive for methadone
2. Be pregnant as determined by human chorionic gonadotropin (hCG) testing at the index ED visit
3. Have a medical or psychiatric condition that requires hospitalization
4. Opioid administration (excluding BUP) at the index ED visit, prior to enrollment, and COWS remains < 8 during ED stay
5. Be actively suicidal or severely cognitively impaired precluding informed consent
6. Present from an extended care facility (e.g., skilled nursing facility)
7. Require continued prescription opioids for a pain condition
8. Be a prisoner or in police custody at the time of index ED visit
9. Be currently (anytime within the past 14 days) enrolled in formal addiction treatment, including by court order. Patients enrolled in formal addiction who are not receiving MOUD are eligible
10. Be unable to provide reliable locator information including 2 contact numbers in addition to their own
11. Be unwilling to follow study procedures (e.g., unwilling to provide permission to contact referral provider/program or unavailable for the follow-up assessments)
12. Have prior enrollment in the current study component

Ancillary Component:

Inclusion Criteria:

1. Be 18 years or older
2. Treated in the ED during study screening hours
3. Meet DSM-5 diagnostic criteria for moderate to severe opioid use disorder
4. Have a COWS <8
5. Have a urine toxicology test that is positive for opioids (opiates, oxycodone, or buprenorphine). Patients with urines that are only positive for fentanyl on the point of care test strip will be eligible if their clinical history and physical exam are consistent with opioid use and they meet DSM-5 criteria for moderate to severe OUD.
6. Be able to speak English sufficiently to understand the study procedures and provide written informed consent to participate in the study

Exclusion Criteria:

1. Have a urine toxicology test that is positive for methadone
2. Be pregnant as determined by human chorionic gonadotropin (hCG) testing at the index ED visit
3. Have a medical or psychiatric condition that requires hospitalization at the index ED visit, prior to enrollment
4. Be actively suicidal or severely cognitively impaired precluding informed consent
5. Present from an extended care facility (e.g., skilled nursing facility)
6. Require continued prescription opioids for a pain condition
7. Be a prisoner or in police custody at the time of index ED visit
8. Be currently (anytime within the past 7 days) enrolled in formal addiction treatment, including by court order. Patients enrolled in formal addiction treatment but are not receiving MOUD are eligible
9. Be unable to provide reliable locator information including 2 contact numbers in addition to their own
10. Be unwilling to follow study procedures (e.g., unwilling to provide permission to answer daily assessments until day 7)
11. Have prior enrollment in the current study

In line with the National Institutes of Health Helping to End Addiction Long-term (NIH HEAL) Initiative Public Access and Data Sharing Policy, publications and underlying primary data will be made available to the public.

Data will be made available after 1) the primary paper has been accepted for publication, or 2) the data is locked for more than 18 months, whichever comes first.