Journal Logo

Original Research

Emergency Department-initiated Buprenorphine and Referral to Follow-up Addiction Care: A Program Description

Regan, Susan PhD; Howard, Sydney MS; Powell, Elizabeth MPH; Martin, Alister MD, MPP; Dutta, Sayon MD, MPH; Hayes, Bryan D. PharmD; White, Benjamin A. MD; Williamson, Dawn DNP; Kehoe, Laura MD, MPH; Raja, Ali S. MD; Wakeman, Sarah E. MD

Author Information
doi: 10.1097/ADM.0000000000000875

Abstract

Initiating treatment for patients with opioid use disorder (OUD) with buprenorphine in the emergency department (ED) improves retention in addiction treatment and reduces self-reported opioid use.1,2 However, a minority of ED clinicians report a high level of readiness to start buprenorphine treatment, due in part to concerns about adequate linkage to ongoing addiction care.3,4 Models for successful implementation of ED-initiated buprenorphine have generally incorporated rapid follow-up within 24 to 72 hours of ED discharge; however, this may be difficult in real-world practice.1–5 In addition, clinical decision support and education may increase ED provider initiation of buprenorphine for patients with OUD.6

The Massachusetts General Hospital (MGH) Bridge Clinic is a low-threshold, outpatient service that provides on-demand, immediate access to addiction treatment including medication for opioid use disorder (MOUD). It offers daily walk-in services, enabling patients discharged from the hospital or ED to immediately access ongoing care, including MOUD. In addition to MOUD, the Bridge Clinic offers care for all types of substance use disorder, including pharmacotherapy, group and individual counseling, recovery coaching, resource support, harm reduction services, and psychiatric care for co-occurring mental illness. In 2018, MGH launched a behavioral economics informed campaign called “Get Waivered” to encourage emergency medicine physicians to obtain a drug enforcement agency x-waiver to prescribe buprenorphine. This effort resulted in 95% of attending physicians in the ED being able to prescribe buprenorphine.7 However, a subsequent evaluation found that although 80% of providers in the ED felt buprenorphine should be prescribed to ED patients, only 44% felt personally prepared to discuss treatment with patients.4 Protocol support with the electronic health record (EHR) and a robust referral system for outpatient follow-up were 2 supports identified by providers as interventions that might increase their comfort with providing MOUD.4

To increase the willingness of ED providers to initiate buprenorphine and the number of ED patients with OUD entering addiction treatment, clinical decision support, and an electronic referral mechanism for ED providers to enter an ambulatory referral order to the MGH Bridge Clinic were created in the EHR. Providers were prompted to enter a referral to the Bridge Clinic by clinical decision support in the EHR through a Best Practice Advisory (BPA). As the Bridge Clinic is open 7 days per week, patients were instructed by the ED provider and through written discharge instructions to go there for follow-up the day after discharge from the ED. The Bridge clinic referral order allowed for the generation of a list of patients for Bridge clinic staff to make additional, direct telephonic outreach to patients referred who had not yet shown for a walk-in appointment, to encourage post-ED engagement. In this program description and evaluation, we describe the details of this model of treatment initiation and referral from the ED and investigate predictors of having OUD addressed during an ED visit and engaging in subsequent care within 34 days of ED discharge.

METHODS

Study Setting and Human Subjects Protection

An urban, university-affiliated tertiary-care hospital ED and Bridge Clinic. This study was deemed to be a Quality Improvement initiative by the Partners Healthcare Institutional Review Board and as such was not formally supervised by the Institutional Review Board per their policies.

Best Practice Advisory

To support the referral process from ED clinicians to the Bridge Clinic, we created a novel BPA in the hospital's EHR (Epic, Verona, WI). The BPA was designed to identify patients with OUD who were not already engaged in treatment. The BPA triggered when an OUD-related diagnosis code was present on the patient's problem list or OUD was documented in the social history section, but not if methadone or buprenorphine were already on the patient's routine medication list or there was no recent history of Bridge Clinic visit notes at our hospital. The BPA was designed to prompt clinicians to discuss the diagnosis and management of OUD with their patients and to consider ordering a referral to the Bridge Clinic (see Appendix for further details, https://links.lww.com/JAM/A281). Clinicians engaged with this BPA in the form of a passive alert in the note creation and disposition sections of the medical record.

Referral to Bridge Clinic

ED providers were able to enter an ambulatory referral order to the Bridge Clinic in the EHR. Although ED providers instructed patients during the ED encounter to go to Bridge Clinic after discharge, to add an additional layer of outreach the Bridge Clinic order created a list of patients which was reviewed by patient care staff at the Bridge Clinic who made subsequent phone calls to patients to encourage them to come in for their visit.

Buprenorphine Initiation

Buprenorphine could be initiated in 2 different ways for patients seen in the ED. Patients could be given a dose of buprenorphine during the ED visit, which was considered buprenorphine initiation in the ED, and then directed to follow-up with the Bridge Clinic. Patients could also be given a buprenorphine dose pack at the time of ED discharge, which contained four to 6 buprenorphine/naloxone 8 mg/2 mg films, instructions for unobserved induction, an intranasal naloxone kit, and information instructing the patient to go to the MGH Bridge Clinic the following day. Patients could receive either of these options for buprenorphine or both, at the discretion of the ED provider.

Study Population

Patients with OUD aged 18 years who visited the ED from January 1st, 2019–December 31st, 2019 were eligible for inclusion. We considered patients to have OUD if they received a diagnosis of OUD or opioid-related overdose at the ED visit, were identified by the BPA as having OUD, or met previously validated criteria for OUD in the past 12 months.8 Visits with a disposition of transfer to another institution or death were excluded. If a patient had >1 eligible visit during the study period, the only first occurring one was retained for analysis.

Data Collection

We obtained demographics, ED visit characteristics including referral status and diagnoses, buprenorphine administration or prescription, OUD treatment history, and completed visit with the Bridge Clinic from the EHR (Epic, Verona, WI).

Outcome Measures

These analyses examine 2 outcomes, addressing OUD in the ED and subsequent treatment engagement. OUD was considered addressed if a patient was referred to Bridge Clinic, was initiated on buprenorphine in the ED or received a home buprenorphine induction kit.

We used the Healthcare Effectiveness Data and Information Set (HEDIS) definition of treatment engagement: an initial encounter within 14 days of discharge, and either 2 subsequent encounters or a subsequent buprenorphine prescription within 34 days of the initial encounter.9 A visit was considered an eligible encounter if it was a Bridge Clinic visit or if was a primary care or mental health outpatient visit that included a billing code for OUD.

Statistical Analysis

We present descriptive statistics for patients seen in the ED during the study period and classified as having OUD. We assessed predictors of the main outcomes, OUD being addressed during the ED visit and subsequent treatment engagement, in generalized linear models using the Poisson distribution. Only the patient's first ED visit during the study period was included in these analyses. The models included age, sex, race, insurance type (Medicaid, Medicare, private or other), ED encounter type (ED only, admitted from the ED or held for observations), discharge disposition (to home or left against medical advice), current opioid OD diagnosis (at the ED visit), current OUD diagnosis, a prescription for buprenorphine in the past 90 days, previous visit to the Bridge Clinic and previous inpatient consult with the Addiction Consult Team (ACT) at MGH. The model predicting subsequent treatment engagement included a term for OUD being addressed during the ED visit and omitted the term for discharge disposition.

We further explored differences in likelihood of OUD being addressed during an ED visit by race/ethnicity. We included all visits in which the patient was classified as having an OUD and used generalized estimating equation models for panel data in these repeated measures analyses. We ran a model predicting OUD being addressed in a model that included race/ethnicity and the other factors in the main outcome model to assess the association of race/ethnicity with the likelihood of receiving treatment. To estimate the size of this effect in terms of the number and proportion of patients not treated we ran a “race-agnostic” model, that is, the same model but dropping the race/ethnicity factor. We used the latter model to calculate the predicted probability that each patient would receive treatment based on the remaining factors. We estimated the predicted number of referrals within each race/ethnicity category by multiplying the mean predicted probability within each category by the number of visits. We present the difference between the observed and predicted number of referrals as a percent of the predicted referrals (100×[observed-predicted]/predicted).

We present adjusted rate ratios (ARR) and 95% confidence intervals (CI). All analyses were conducted using Stata statistical software (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC).

RESULTS

Study Population, Patient Demographics, and Care During the ED Visit

During 2019, there were 1946 patients who met criteria for OUD who completed 4392 ED visits. Patients were predominantly White (81%), male (66%), and publicly insured by Medicaid/Medicare (84%). Median age was 40 years (IQR: 33–51 years). A minority (14%) had received a buprenorphine prescription in the previous 90 days. Most patients seen were discharged from the ED (57%); however, 34% were admitted to the hospital, 9% admitted for observation only, and 13% left prematurely (“against medical advice”) (Table 1).

TABLE 1 - Demographics and Clinical Characteristics
Characteristic N Percent
Male 1283 66
Female 663 34
Age (yr)
 18–29 293 15
 30–39 679 35
 40–49 409 21
 50–64 463 24
 > = 65 102 5
Race/Ethnicity
 Black 148 8
 White 1575 81
 Hispanic/Latinx 129 7
 Asian 4 <1
 Other race/ethnicity 47 2
 Unknown 43 2
Insurance
 Medicaid 1183 61
 Medicare 448 23
 Private 227 12
 Other 88 5
Encounter type
 ED visit only 1114 57
 Held for observation 171 9
 Inpatient admission 661 34
Left against medical advice 261 13
SUD diagnoses
 Opioid use disorder 910 47
 Opioid overdose 137 7
 Alcohol 314 16
 Cocaine 219 11
 Stimulants 86 4
 Sedatives 51 3
 Polysubstance 157 8
Buprenorphine prescription in prior 90 days 277 14
BPA displayed during ED visit 521 27
Prior Bridge Clinic visit 243 12
Prior ACT consult 576 30
Total 1946 100
ACT, Addiction Consult Team; BPA, best practice advisory; ED, emergency department; SUD, substance use disorder.
SUD diagnoses based on discharge diagnoses associated with ED visit.

Each patient's first visit during the study period was retained for the main analyses. The BPA was displayed in 521/1946 (27%) visits (Fig. 1). OUD was addressed in the ED for 298/1946 patients (15%). The rate was higher when the BPA was displayed during the visits (22% vs 13%, P < 0.001). A total of 156/1946 (8%) patients met HEDIS criteria for treatment engagement, meaning they had an initial encounter within 14 days of ED discharge, and either 2 subsequent encounters or a subsequent buprenorphine prescription within 34 days of the initial encounter. Patients were more likely to be engaged after visits during which their OUD was addressed (46/298, 15%) than when it was not (110/1648, 7%; P < 0.001).

F1
FIGURE 1:
OUD treatment and engagement pathways. Best practice advisory is displayed in the electronic medical record during the ED visit for patients identified as having untreated opioid use disorder who are not expected to be admitted (see Appendix for details). OUD may be addressed in the ED by 1 or more of: initiation of buprenorphine during the visit, receipt of a home induction kit, or referral to the Bridge Clinic. Treatment engagement is defined as an initial encounter within 14 days of discharge, and either 2 subsequent encounters or a subsequent buprenorphine prescription within 34 days of the initial encounter. ED, emergency department; OUD, opioid use disorder.

Addressing OUD During the ED Visit

OUD was addressed by referral to the Bridge Clinic for 207 (11%) patients, by buprenorphine initiation during the visit for 106 (5%), and by receipt of buprenorphine home induction packs for 56 (3%) (Fig. 2 Panel A). Some patients (N = 63) received 2 or more of these interventions (Supplementary Figure 1, https://links.lww.com/JAM/A282). The number and percent of patients whose OUD was addressed is shown by demographic and clinical characteristics in Table 2. In unadjusted analyses, patients were more likely to have their OUD addressed if they were male or the BPA was displayed and less likely if they were ≥40 years old compared to those under 30 years, had Medicare or commercial insurance (vs Medicaid), were admitted, left against medical advice or had a previous ACT consultation. In the multivariate model, display of the BPA was associated with greater likelihood of addressing OUD, while being age 50 years or older, admitted, leaving against medical advice, and prior ACT consultation were associated with lower rates of addressing OUD in the ED (Table 2).

F2
FIGURE 2:
OUD treatment and treatment engagement outcomes. Panel A: Frequency of measures to address OUD during the ED visit. The bars represent the number among the total 1946 patients who were referred to the Bridge Clinic from the ED (darker gray), were initiated on buprenorphine during the ED visit (dark gray), received a home induction kit to take home (medium gray), or received any of the 3 interventions (light gray). Panel B: Frequency of meeting HEDIS criteria for treatment engagement. The bars represent the number among the total 1946 patients with a clinic visit within 14 days of the ED visit (darker gray), the subset of that number who had 2 additional visits within 34 days of the ED visit (dark gray), and the number who received a buprenorphine prescription within 34 days of the ED visit (medium gray). To meet HEDIS criteria for engagement (light gray), the patient must have had a clinic visit within 14 days plus either 2 more visits or a prescription. ED, emergency department; OUD, opioid use disorder.
TABLE 2 - Multivariable Models Predicting OUD Being Addressed During ED Visit and Subsequent Engagement in Treatment (N = 1946)
OUD Addressed in ED Treatment Engagement
N % RR 95% CI ARR 95% CI N % RR 95% CI ARR 95% CI
Male 215 17 1.34 (1.04–1.72) 1.29 (0.99–1.66) 113 9 1.36 (0.96–1.93) 1.35 (0.95–1.94)
Female 83 13 1.00 Ref. 1.00 Ref. 43 6 1.00 Ref. 1.00 Ref.
Age (yr)
 18–29 64 22 1.00 Ref. 1.00 Ref. 27 9 1.00 Ref. 1.00 Ref.
 30–39 113 17 0.76 (0.56–1.03) 0.82 (0.60–1.12) 71 10 1.13 (0.73–1.77) 1.12 (0.71–1.75)
 40–49 59 14 0.66 (0.46–0.94) 0.77 (0.54–1.10) 26 6 0.69 (0.40–1.18) 0.68 (0.39–1.17)
 ≥50 62 11 0.50 (0.35–0.71) 0.68 (0.47–0.98) 32 6 0.61 (0.37–1.03) 0.76 (0.44–1.33)
Race/Ethnicity
 Black 18 12 0.78 (0.48–1.26) 0.62 (0.38–1.00) 7 5 0.55 (0.26–1.18) 0.72 (0.33–1.56)
 White 245 16 1.00 Ref. 1.00 Ref. 135 9 1.00 Ref. 1.00 Ref.
 Hispanic/Latinx 20 16 1.00 (0.63–1.57) 0.84 (0.53–1.34) 6 5 0.54 (0.24–1.23) 0.51 (0.22–1.16)
 Other/Unknown race 15 15 1.02 (0.61–1.73) 0.92 (0.54–1.56) 8 8 0.99 (0.49–2.03) 0.88 (0.43–1.80)
Insurance
 Medicaid 199 17 1.00 Ref. 1.00 Ref. 99 8 1.00 Ref. 1.00 Ref.
 Medicare 53 12 0.70 (0.52–0.95) 0.97 (0.70–1.33) 30 7 0.80 (0.53–1.20) 1.06 (0.68–1.65)
 Commercial 25 11 0.65 (0.43–0.99) 0.66 (0.43–1.01) 23 10 1.21 (0.77–1.91) 1.38 (0.87–2.21)
 Other insurance 21 24 1.42 (0.90–2.22) 1.07 (0.67–1.68) 4 5 0.54 (0.20–1.48) 0.72 (0.26–1.99)
BPA displayed 115 22 1.72 (1.36–2.17) 1.30 (1.01–1.68)
Admitted as inpatient 22 3 0.15 (0.10–0.24) 0.18 (0.12–0.28)
Left against medical advice 24 10 0.57 (0.38–0.87) 0.57 (0.37–0.87) 15 6 0.70 (0.41–1.19) 0.76 (0.44–1.31)
Past Bridge Clinic encounter 43 18 1.18 (0.86–1.63) 1.33 (0.92–1.91) 58 24 4.15 (3.00–5.74) 2.12 (1.44–3.11)
Past ACT consultation 59 10 0.59 (0.44–0.78) 0.65 (0.48–0.88) 51 9 1.16 (0.83–1.61) 0.98 (0.69–1.40)
Buprenorphine prescription in past 90 d 38 14 0.88 (0.63–1.24) 0.76 (0.52–1.10) 75 27 5.58 (4.08–7.64) 4.14 (2.89–5.92)
OUD addressed in ED 46 15 2.31 (1.64–3.26) 2.30 (1.62–3.27)
Total 298 15 156 8
ACT, Addiction Consult Team; ARR, adjusted rate ratio; BPA, best practice advisory; CI, confidence interval; ED, emergency department; OUD, opioid use disorder.
Referred to Bridge Clinic, buprenorphine initiated during ED visit, or received buprenorphine home induction kit.
Had an encounter within 14 days of ED visit plus (2 subsequent visits or a prescription for buprenorphine) within the next 34 days.

Black patients had lower rates of having their OUD addressed (ARR: 0.62, 95% CI: 0.38–1.00), but this difference only approached statistical significance (P = 0.053), possibly due to the small proportion of Black patients (148/1946). We explored this further in a repeated measures analysis that included all eligible visits for all patients during the study period. This analysis included all 4392 ED visits by 1946 patients, including 372 visits by 148 Black patients, and included the same predictors as the previous model. In this analysis, OUD was approximately half as likely to be addressed during an ED visit if the patient was Black as when the patient was white (ARR: 0.46, 95% CI: 0.30–0.72, P = 0.001). To estimate the proportion of patients who were not treated due to this association, we repeated the model with the race/ethnicity terms removed, then calculated the predicted probability that OUD would be addressed at the visit with the remaining variables and compared that to the observed frequencies by race (Fig. 3). The model without race predicted that OUD would be addressed in 55/372 visits by Black patients, but in fact, it was addressed in only 28 visits, 49% ((55–28)/55) fewer than expected. A similar but smaller discrepancy was observed for visits by Hispanic/Latinx patients: the model without race predicted OUD would be addressed during 45/308 visits by Hispanic/Latinx patients, but it was addressed in only 34 visits, 25% ((45–34)/45) fewer than expected.

F3
FIGURE 3:
Observed versus predicted OUD treatment during ED visit by race/ethnicity of patient. Predicted treatment rates based on a repeated measures model (4392 visits by 1946 patients) controlling for demographic (sex, age, insurance status) and clinical (best practice advisory displayed, (inpatient admission, discharge against medical advice, prescription for buprenorphine within 90 days, past Bridge Clinic visit or inpatient addiction consult) characteristics but not race/ethnicity. A negative value indicates fewer patients were treated than expected by this model that ignores race/ethnicity. ED, emergency department; OUD, opioid use disorder.

Subsequent Treatment Engagement

HEDIS criteria for treatment engagement were met by 156 patients (8%, Fig. 2). The required initial OUD encounter was completed by 237 patients within 14 days of discharge from the ED, 122 of whom satisfied engagement criteria by completing 2 more visits within 34 days. A total of 207 patients received a buprenorphine prescription within 34 days of ED discharge; 96 also had the required initial encounter and thereby satisfied engagement criteria (Supplementary Figure 2, https://links.lww.com/JAM/A283).

In the multivariate model, patients who had previously engaged in the Bridge Clinic, who had received buprenorphine in the past 90 days, or whose OUD was addressed during the ED visit were more likely to engage in OUD care following ED discharge (Table 2).

DISCUSSION

In this program evaluation, we found that having OUD addressed during an ED encounter either through buprenorphine initiation or a referral to Bridge Clinic was associated with a greater likelihood of subsequent treatment engagement in OUD care. In this real-world evaluation, among the 8% of patients who met strict criteria for treatment engagement following discharge, nearly two thirds had received a subsequent prescription for buprenorphine emphasizing the importance of MOUD in ensuring continued treatment retention. These findings suggest that buprenorphine initiation and Bridge Clinic referral are important components of successful ED treatment for patients with OUD.

The patients who had their OUD addressed in the ED in this program were predominantly White men. Patients who were Black or Hispanic/Latinx were referred less often than predicted, raising concern that racism may impact referral rates. Prior research has demonstrated racial inequities in the receipt of addiction treatment, particularly in access to buprenorphine.10–13 In addition, studies show the presence of implicit bias among physicians, including in emergency medicine.14,15 Although the impact of this on clinical decision-making related to addiction treatment referral in the ED has not been tested, research has demonstrated racial disparities in pain treatment in the ED.16 Further research is needed to evaluate the impact of racism on provider decision-making related to care for patients with OUD in the ED, to assess patient experiences of racism in OUD care, and to develop culturally competent interventions within an antiracism framework.

Despite the evidence that buprenorphine initiation in the ED is more effective than brief intervention or referral to treatment, only 5% of patients with OUD were initiated on buprenorphine in the ED in this study and 3% were given home dose packs. This is in line with prior evaluations examining ED buprenorphine initiation. For example, 1 study found a baseline rate of buprenorphine initiation of 3.5% and a subsequent increase to 6.6% following a clinical decision support intervention.4 Our slightly higher rate of ED buprenorphine initiation combined with dose pack prescribing may be due to prior efforts at our institution to increase ED provider confidence with OUD treatment7 as well as the existence of our Bridge Clinic which allowed timely follow-up and may have enabled ED providers to feel more comfortable prescribing. Still, few patients were started on buprenorphine despite a majority of ED physicians having an X-waiver to prescribe buprenorphine. This demonstrates an opportunity for improvement and indicates that X-waiver training may not be sufficient to substantially increase prescribing rates. Future research is needed to test interventions to increase the frequency of ED provider prescribing and rigorously evaluate whether X-waiver training helps or hinders access to buprenorphine. Additionally, the ability to refer to post-ED addiction care may not be enough to increase provider comfort with MOUD treatment initiation.

Referring patients to Bridge Clinic occurred more often than initiating buprenorphine in the ED. Several possibilities exist for why this might be. First, the time pressure of needing to move patients out of the ED may mean providers defer buprenorphine initiation to the Bridge Clinic. Second, there may be a perception that a “warm hand-off” and connecting a patient same day to the Bridge Clinic is better than prescribing buprenorphine and instructing them to follow-up in the Bridge Clinic. Third, this may indicate ongoing discomfort with the mechanics of initiating buprenorphine among ED providers. Given prior studies demonstrating the importance of buprenorphine initiation during the ED encounter as a critical ingredient in retention on MOUD, further implementation work is needed to increase ED-based buprenorphine treatment.1 The strong association between buprenorphine prescription in the 90 days before ED visit and subsequent treatment engagement indicates the importance of MOUD in retaining people in care, even after the potential disruption of whatever acute medical event resulted in an ED encounter.

Patients were more likely to have their OUD addressed in the ED when the BPA was displayed, suggesting that clinical decision support in the EHR may help increase treatment initiation. However, among 1946 patients with OUD, the BPA fired only for 27% indicating that the BPA criteria may need modification. The BPA was designed to be specific rather than sensitive to minimize alert fatigue for providers and only fire for patients who were unlikely to be admitted and were not recently engaged in OUD treatment. Given the overall low rates of treatment initiation and the seeming benefit of the BPA in prompting providers to offer treatment, creating more inclusive criteria may lead to better clinical outcomes.

The role of patient preference for type of MOUD treatment is another important factor that has not been examined in ED-initiated OUD care. Other studies have explored the importance of patient prior treatment experiences and preferences in OUD treatment selection, notably indicating that some patients prefer methadone treatment to buprenorphine.17 Expanding MOUD options in the ED to include methadone initiation and direct linkage to an opioid treatment program may increase the uptake of treatment overall.

Lastly, the low overall subsequent engagement rates emphasize the need for ongoing innovation to ensure patients with OUD are able to receive needed treatment after ED discharge. Although our model utilizes peer recovery coaches and assertive telephonic outreach about Bridge Clinic follow-up, greater intensity interventions may be needed.

This program evaluation was subject to at least 4 limitations. First, we classified patients as having OUD based on available data in the EHR and it is possible we either over or under identified OUD. However, we utilized an algorithm for identifying OUD which has been previously validated. Second, because we were limited by lack of data outside of our system it is possible that some patients connected successfully to addiction treatment elsewhere including patients who may have been appropriately connected to methadone treatment. Subsequent research using claims data could more thoroughly evaluate treatment engagement in any venue. Third, we were not able to capture whether patients were offered OUD care in the ED, only whether they received it, so it is possible that a larger number were offered it and some declined. For example, in a clinical trial of buprenorphine initiation among 329 ED patients, 171 eligible patients declined to participate.1 Lastly, we did not account for provider level factors such as the experience of the provider or comfort with OUD care. Ongoing research is needed to evaluate how to change provider behavior in providing OUD care.

CONCLUSIONS

In this program evaluation, addressing OUD in the ED through buprenorphine initiation and Bridge Clinic referral was associated with improved treatment engagement. Black and Hispanic/Latinx patients received OUD care less often than expected.

REFERENCES

1. D’Onofrio G, O’Connor PG, Pantalon MV, et al. Emergency department-initiated buprenorphine/naloxone treatment for opioid dependence: a randomized clinical trial. JAMA 2015; 313 (16):1636–1644.
2. Strayer RJ, Hawk K, Hayes BD, et al. Management of opioid use disorder in the emergency department: a white paper prepared for the American Academy of Emergency Medicine. J Emerg Med 2020; 58 (3):522–546.
3. Hawk KF, D’Onofrio G, Chawarski MC, et al. Barriers and facilitators to clinician readiness to provide emergency department-initiated buprenorphine. JAMA Netw Open 2020; 3 (5):e204561.
4. Im DD, Chary A, Condella AL, et al. Emergency department clinicians’ attitudes toward opioid use disorder and emergency department-initiated buprenorphine treatment: a mixed-methods study. West J Emerg Med 2020; 21 (2):261–271.
5. Bogan C, Jennings L, Haynes L, et al. Implementation of emergency department-initiated buprenorphine for opioid use disorder in a rural southern state. J Subst Abuse Treat 2020; 112S:73–78.
6. Holland WC, Nath B, Li F, et al. Interrupted time series of user-centered clinical decision support implementation for emergency department-initiated buprenorphine for opioid use disorder. Acad Emerg Med 2020; 27:753–763.
7. Martin A, Kunzler N, Nakagawa J, et al. Get waivered: a resident-driven campaign to address the opioid overdose crisis. Ann Emerg Med 2019; 74 (5):691–696.
8. Wakeman SE, Larochelle MR, Ameli O, et al. Comparative effectiveness of different treatment pathways for opioid use disorder. JAMA Netw Open 2020; 3 (2):e1920622.
9. National Committee on Quality Assurance (NCQA). NCQA Updates Quality Measures for HEDIS 2018. Available at: https://www.ncqa.org/news/ncqa-updates-quality-measures-for-hedis-2018/. Accessed January 31, 2021.
10. Stein BD, Dick AW, Sorbero M, et al. A population-based examination of trends and disparities in medication treatment for opioid use disorders among Medicaid enrollees. Subst Abus 2018; 39 (4):419–425.
11. Goedel WC, Shapiro A, Cerdá M, et al. Association of racial/ethnic segregation with treatment capacity for opioid use disorder in counties in the United States. JAMA Netw Open 2020; 3 (4):e203711.
12. Lagisetty PA, Ross R, Bohnert A, et al. Buprenorphine treatment divide by race/ethnicity and payment. JAMA Psychiatry 2019; 76 (9):979–981.
13. Hansen H, Siegel C, Wanderling J, et al. Buprenorphine and methadone treatment for opioid dependence by income, ethnicity and race of neighborhoods in New York City. Drug Alcohol Depend 2016; 164:14–21.
14. Dehon E, Weiss N, Jones J, et al. A systematic review of the impact of physician implicit racial bias on clinical decision making. Acad Emerg Med 2017; 24 (8):895–904.
15. Samuels EA, Boatright D, Sanchez LD, et al. Clinical vignettes inadequate to assess impact of implicit bias: concerning limitations of a systematic review. Acad Emerg Med 2017; 24 (12):1531–1532.
16. Benzing AC, Bell C, Derazin M, et al. Disparities in opioid pain management for long bone fractures. J Racial Ethn Health Disparities 2020; 7:740–745.
17. Yarborough BJ, Stumbo SP, McCarty D, et al. Methadone, buprenorphine and preferences for opioid agonist treatment: a qualitative analysis. Drug Alcohol Depend 2016; 160:112–118.
Keywords:

Bridge Clinic; buprenorphine; emergency department; opioid; racism

Supplemental Digital Content

Copyright © 2021 American Society of Addiction Medicine