a b s t r a c t

Objectives: Opioid use disorder (OUD) is a national epidemic, and Black and Hispanic patients are less likely to re- ceive treatment when compared to white patients. In this study, race was used as a proxy to assess potential ef- fects of racism on the referral process for OUD treatment. Our primary aim was to examine whether Black or Hispanic patients experienced increased barriers to inpatient OUD detoxification (detox) placement at a community-integrated, substance use disorder support program based in an emergency department (ED). Our secondary aim was to determine if Black and Hispanic patients were more likely to have >3 referrals.

Methods: This retrospective cohort study was conducted at a large urban safety-net hospital and included pa- tients seen in the ED from July 2018 to September 2019 with ICD-10 codes for an opioid-related visit and who sought placement to inpatient detox. A generalized linear mixed model controlling for multiple visits, age, sex, insurance, time, day of week, and time of year was used to assess the association between race/ethnicity and hy- pothesized barriers to placement. The proportion of patients with >3 visits for referral to inpatient detox was compared between Black and Hispanic patients and white patients using a chi-squared test.

Results: We identified 1733 encounters from 782 unique patients seeking connection to inpatient detox for OUD. Of the 1733 encounters, 45% were among Black and Hispanic patients. Hispanic and Black men had significantly lower odds of having a barrier to inpatient OUD detox than white men (OR = 0.734, 95% CI 0.542-0.995). No sig- nificant difference was found for Hispanic and Black women (OR = 1.212, 95% CI 0.705-2.082). More Black and Hispanic patients experienced >3 referrals to inpatient detox compared to white patients (19.2% vs 12.9%, p = 0.016).

Conclusions: This study suggests in the context of near-universal health insurance coverage, an ED-based OUD support program staffed by diverse community members can mitigate inequities in access to inpatient detox. However, the increased number of ED visits for OUD detox placement by Black and Hispanic patients suggests racial inequities in OUD treatment exist after linkage to care. Additional research should explore the causes, spe- cifically structural and interpersonal racism, and determine solutions to address racial inequities in detox place- ment as well as maintenance in treatment programs.

Published by Elsevier Inc.

1. Introduction

? Prior Presentations (Virtual due to COVID-19):Society for Academic Emergency Medicine Annual MeetingLightning Oral PresentationMay 15, 2020Society for Academic Emergency Medicine Annual MeetingPoster PresentationMay 15, 2020

* Corresponding author at: West Los Angeles Veterans Affairs, National Clinician Scholars Program, 1100 Glendon Ave., Ste. 900, Los Angeles, CA 90025, United States of America.

E-mail addresses: [email protected] (J. Faiz), [email protected] (E. Bernstein), [email protected] (J.N. Dugas), [email protected] (E.M. Schechter-Perkins), [email protected] (L. Nentwich), [email protected] (K.P. Nelson), [email protected] (E.C. Cleveland Manchanda), [email protected] (L. Young), [email protected] (J.R. Pare).

Opioid use disorder afflicts an estimated 16 million peo- ple worldwide, with over 120,000 deaths annually [1]. In the United States, 128 people die of opioid overdose daily, making it the most lethal drug epidemic in American history [2]. Because in- dividuals with OUD are often socioeconomically and functionally marginalized, the emergency department (ED) is often a primary access point to the healthcare system and thus disproportionately cares for these patients [3]. For this reason, ED visits are an impor- tant opportunity to connect patients with OUD to counseling and treatment [4].

https://doi.org/10.1016/j.ajem.2022.01.037 0735-6757/Published by Elsevier Inc.

While deaths have decreased since the opioid epidemic was declared a public health emergency in 2017, rates of opioid fatalities have increased among Black and Hispanic people [5,6]. During the COVID-19 pandemic, overdose deaths among Black individuals increased by more than 50% compared to white counterparts in some communities [7,8]. This may be due to delays in linkage to OUD treat- ment, decreased access to buprenorphine treatment in communities of color, and a greater likelihood of inadequate methadone dosing for black patients [9-11]. In Massachusetts, the odds of receiving OUD treat- ment following hospitalization for overdose is 29% lower for Black patients compared to white patients, even after adjusting for socio- demographic characteristics [12]. This raises many concerns, specifically how structural and interpersonal racism may influence the ability of marginalized and minoritized individuals to successfully engage with treatment for OUD [13].

With the hope of pinpointing where hurdles to OUD treatment lie, the primary aim of this investigation was to examine whether being Black and/or Hispanic was associated with barriers to inpatient detox placement among patients who presented to our substance use disorder support program, Project ASSERT (Alcohol & Substance use Services, Education and referral to treatment), to discuss treat- ment for their substance use disorder having also received ED care for an OUD-related healthcare problem. Our secondary aim was to assess if Black and Hispanic patients within this cohort were more likely to have recurrent visits for OUD inpatient detox placement (>3) than white patients. Given previously reported racial dispar- ities, we hypothesized that Black and/or Hispanic patients would be more likely to have barriers to treatment, and would be more likely to seek care multiple times.

1. Methods
   1. Study design and sample

This is a retrospective cohort study conducted at an academic urban safety net hospital between July 2018 - October 2019. Participants included all patients >=18 years old who presented to the ED for an OUD-related diagnosis who received a referral from Project ASSERT to inpatient OUD detox. This study was reviewed and determined to be exempt by our hospital’s Institutional Review Board.

• 1. Study setting and population

This study was conducted at the busiest ED in the region, serving more than 130,000 patients per year. Approximately 57% of patients are from under-served populations and 32% of patients do not speak English as a primary language.

• • 1. Project ASSERT

Since 1994, Project ASSERT has offered alcohol and drug use assess- ments and linkage to treatment for patients receiving care in this ED. Project ASSERT is staffed by licensed alcohol and drug counselors (LADC) from diverse backgrounds, with 78% of staff self-identifying as Black or Hispanic. Patients who present to the ED for OUD-related com- plaints, as well as those seeking detox services related to OUD, are re- ferred for consultation with an LADC from Project ASSERT. These patients are then either met in the ED by a LADC or self-present to the adjacent Project ASSERT office after discharge from the ED. Project ASSERT’s hours at the time of this study were 8 am to 11 pm daily, in- cluding weekends and holidays. Pregnant women requesting OUD treatment are admitted to the hospital for care, and therefore are not referred for treatment.

The LADC performs an intake history, which consists of the identifi- cation of a substance use disorder, consent for treatment, discussion of options and resources, active referral or linkage with support services, and follow-up [14]. The LADC also completes an extensive drug use

and psychosocial assessment during the encounter, which is recorded in the hospital electronic medical records (EMR). If a patient requests placement in an acute treatment facility (inpatient detox), the LADC contacts various facilities, supports patients during a treatment center telephone intake, and completes a warm handoff to the treatment facil- ity. Patient eligibility for inpatient detox is determined by factors such as patient preference, substance use history and psychosocial assessment, and bed availability. successful placement is then dependent on the patient being accepted by a detox facility.

In the greater Boston area, there are four main inpatient OUD detox facilities, all of which accept Medicaid insurance. One facility has 50 beds and another has 32 beds, all of which are designated male beds. Another 30-bed facility has 9 beds reserved for females, and there is a 14-bed facility which can accommodate both sexes. The LADC provides Uber Health (San Francisco, CA) transportation to detox facilities with- out cost to the patient.

• 1. Study protocol

As a part of ongoing hospital quality efforts to improve care for patients with OUD, two datasets were generated for administrative purposes. The first dataset contains OUD-relatED patient visits to the ED, including visits for overdose, from October 1, 2018 through September 30, 2019. This dataset was created by abstraction of charts from EMR based on ICD-10 codes (Appendix A). The second dataset was a prospective administrative Project ASSERT dataset consisting of patient information regarding detox referrals, barriers to access and placements from July 1, 2018 through October 31, 2019. Participants were required to receive care in both the ED and Project ASSERT during the study period for inclusion. Participants’ visits to Project ASSERT did not necessarily immediately follow their ED visit. We included patients within these time frames based on what was available in the administrative dataset as well as to allow for delayed office visits after the ED visit if the patient sought detox later when personally ready.

Patients’ medical record numbers were used to link patient visits between datasets. To meet inclusion criteria, patients had to have both an OUD-related visit in the ED and at least one visit to Project ASSERT during which a referral to inpatient detox was attempted. Referral to inpatient detox was defined as: securing a bed at the facil- ity and documenting the placement in the EMR, or if a barrier is doc- umented in the EMR, signifying that the patient was referred but a bed was not secured. Patients missing race, ethnicity or age were ex- cluded from analysis. Our investigation focused on our three largest patient populations, those who identify as white, Black, and/or His- panic. We excluded missing and other non-Hispanic racial groups due to concerns that small numbers would lead to inaccurate conclu- sions. Patients who chose to initiate medication for OUD (MOUD) at the low-barrier outpatient clinic in lieu of detox referral were not included in the study.

Notably, in this study, we utilized race as a proxy to assess po- tential effects of racism on the referral process for OUD treatment [15]. In this study, race/ethnicity were classified as non-Hispanic Black, non-Hispanic white, and Hispanic (patients who identified as white and Hispanic, Black and Hispanic, Hispanic listed as race, and other or unknown race with Hispanic listed as ethnicity). Age, sex, race and ethnicity were collected by trained registration staff based on patient self-report upon presentation. Referral informa- tion including date and time of visit were obtained, allowing for comparisons based on time, day, and season of visit (winter, spring, summer, fall).

All data were obtained from the two datasets, with the exception of insurance status which was obtained through the hospital’s Clinical Data Warehouse (1UL1TR001430) by medical record number based on date of ED visit [16]. Insurance status was categorized as commercial, Medicaid, Medicare, Massachusetts free care, or other.

• 1. Outcome measures

When a referral to inpatient detox is attempted and not successful, LADCs document barriers to a patient’s placement. The potential bar- riers available for selection via a drop-down menu by LADCs include: no bed available, barred, did not have outpatient prescription medica- tions, recently left detox, no identification or green card, no insurance or insurance problems, or other. Notably, detox facilities may require patients to have prescription medications in-hand prior to placement. More than one barrier can be selected for each attempt at detox place- ment. If any barrier was listed, the patient was classified as having not been placed. If no barrier was listed and the name of the inpatient detox facility at which the participant was placed was listed, this indi- cated the patient was placed at an inpatient detox center.

• 1. Data analysis

To assess racial inequities in barriers to detox placement, a logistic generalized linear mixed model (GLMM) was used to assess the associ- ation between the combined cohort of Black and Hispanic patients and the dichotomous outcome of any identified barrier to inpatient detox placement compared with non-Hispanic whites. white race was used as the reference group given white patients have been found to have significantly higher odds of treatment for OUD than Black and Hispanic patients in Massachusetts [12]. The GLMM was selected for this analysis as some patients had more than one visit and the model accounts for the clustering of multiple visits for some patients by including a random ef- fect for each patient. A multivariable GLMM model was also conducted, controlling for multiple visits per patient and covariates, including age, sex, insurance, and time, day, and season of visit.

Power calculation was performed prior to conducting our study demonstrating that with an overall acceptance-to-treatment rate of

70%, 1716 observations (50% white and 50% Black or Hispanic) would achieve 80% power at the 0.05 level of significance to detect a 25% de- crease in odds of white patients having a barrier to placement compared to Black and Hispanic patients.

In a secondary analysis, a dichotomous outcome was used to assess for multiple encounters for referral to inpatient detox, defined as more than 3 visits for referral to detox during the study period (event oc- curred) or 3 and fewer visits (event did not occur). Three was selected as a value slightly higher than the national median of 2 recovery at- tempts [17]. A chi-squared test was used to compare the proportion of Black or Hispanic patients with more than 3 visits to that of white pa- tients with more than 3 visits. P-values less than 0.05 were considered significant. Sample size and power calculations were performed using PASS-19 software and analyses were performed using SAS v9.4 [18].

1. Results
   1. Study sample characteristics

A total of 1733 ED encounters for connection to inpatient detox for OUD occurred by 782 unique patients during the study period (Fig. 1). Among the patient encounters, 40.5% were Black or Hispanic and 59.5% were non-Hispanic white. A majority of participants (55%) had one visit to Project ASSERT during the study period; however, 20% of pa- tients had two visits, 9% had three visits, and 15.5% had more than three visits. The mean age of participants referred to inpatient detox for OUD was 48 years old (+-10.7) (Table 1). We excluded 230 encounters of pa- tients who were referred to initiate MOUD in the outpatient setting, of whom 38% were Black or Hispanic and 57% were non-Hispanic white. The proportion of Black and Hispanic encounters compared to white en- counters that received MOUD was not significantly different than those that opted for inpatient detox treatment (p = 0.12).

Fig. 1. Patient selection flow chart.

Patient selection criteria from database of OUD-related patient visits to the ED from October 1, 2018 through September 30, 2019 and Project ASSERT database patient information regard- ing detox referrals and placements from July 1, 2018 through October 31, 2019. Patients not referred to inpatient opioid detox had other primary Substance use disorders or were deter- mined to be better suited for or elected outpatient MOUD.

Table 1

Patient characteristics.

Overall Barrier to placement

No barrier to placement

p-value

Table 3

Odds of barrier to placement to inpatient detox for opioid use disorder

(N = 782)				(N = 339)			(N = 443)				Hispanic & Black non-Hispanic+-	0.734 (0.542, 0.995)	1.212 (0.705, 2.082)
Age (years), mean, std	47.90	10.69	48.40		10.15	47.52		11.08	0.2503?	Age		1.003 (0.989, 1.017)	1.008 (0.979, 1.037)
Race/ethnicity, n, %										4:00 PM-7:59 PM visitEUR		2.191 (1.563, 3.071)	1.154 (0.649, 2.051)
Hispanic &	531	67.90	230		67.85	301		67.95	0.9765+	8:00 PM-6:59 AM visitEUR		6.250 (4.419, 8.838)	2.757 (1.592, 4.773)
Non-Hispanic Black									Weekend visit?			0.999 (0.743, 1.344)	0.939 (0.574, 1.534)
Non-Hispanic white	251	32.10	109		32.15	142		32.05	Fall Visit+			0.786 (0.551, 1.119)	0.866 (0.476, 1.575)
Sex, n, %									Winter Visit+			0.990 (0.666, 1.470)	0.925 (0.467, 1.832)
Female	161	20.59	67		19.76	94		21.22	0.6180+	Spring Visit+		0.917 (0.617, 1.363)	0.846 (0.439, 1.632)
Male	621	79.41	272		80.24	349		78.78		Medicaid?		0.434 (0.213, 0.883)	1.069 (0.188, 6.114)
Insurance, n, %										Medicare?		0.684 (0.310, 1.510)	2.377 (0.371, 15.224)

Males (n = 1306) Females (n = 427) Adjusted OR (95% CI) Adjusted OR (95% CI)

Commercial	42	5.37	19	5.60	23	5.19	0.5443+
Medicaid	604	77.24	262	77.29	342	77.20
Medicare	118	15.09	47	13.86	71	16.03
Free care & other	2	0.26	1	0.29	1	0.23
Missing	16	2.05	10	2.95	6	1.35

Characteristics of patients who presented to the emergency department for an OUD-related diagnosis from 10/1/2018-9/30/2019 and who also had a referral to inpa- tient OUD detox from 7/1/2018-10/31/2019 from Project ASSERT.

? Fisher’s [49] Exact test.

* two-independent samples t-test.

⁺ chi-squared test of independence.

Of the 1733 encounters, 486 encounters documented at least one barrier to inpatient detox placement (28%). A total of 531 barriers were listed. The Most frequently cited barriers for placement were lack of bed availability 345/531 (71%), not having outpatient medica- tions 53/531 (11%) and too recently leaving detox 16/531 (3%). For 102/531 (21%) of patients who were referred to inpatient detox, a bar- rier of “other” was listed (Table 2).

• 1. Outcome measures

There was no significant difference between the Black and Hispanic cohort compared to the white cohort in having one or more barriers to placement for inpatient OUD treatment in the unadjusted model (OR 0.78, 95%CI 0.60-1.00, Table 3). Due to concerns of bed availability based on sex, the interaction between sex and race was evaluated in the adjusted model. Within the multivariable GLMM model there was a statistically significant observed interaction between race and sex (OR = 1.922, 95% CI 1.073-3.443). Therefore, sex was removed from the model and a model was run for each sex separately. Hispanic and Black non-Hispanic men had statistically significantly lower odds of hav- ing a barrier compared with white non-Hispanic men (OR = 0.734, 95% CI 0.542-0.995), while Hispanic and Black non-Hispanic women did not have a statistically significant difference in having a barrier compared with white non-Hispanic women (OR = 1.212, 95% CI 0.480-2.082).

Among the 782 study participants, 121 (15.5%) presented more than 3 times to Project ASSERT and were referred to inpatient detox for OUD.

Table 2

Barriers for inpatient detox for opioid use disorder.

Free Care & Other Insurance^? 0.481 (0.138, 1.673) 0.774 (0.039, 15.220)

^+- Reference value = white, non-Hispanic

^EUR Reference value = 7:00 AM-3:59 PM

* Reference value = weekday

⁺ Reference value = summer

^? Reference value = Commercial insurance

When assessed, 19.2% of Black and Hispanic participants had more than 3 visits while only 12.9% of white participants experienced more than 3 visits (p = 0.016).

1. Discussion

Many patients with OUD self-present to the ED seeking care, includ- ing assistance with placement to inpatient detox facilities. In this cohort of patients with OUD seeking inpatient detox placement via an ED-linked substance use disorder support program, we found Black and Hispanic men to have fewer barriers identified than white patients, and no difference by race for women. Despite fewer identified barriers to placement, Black and Hispanic patients were more likely to have repeat visits for referral to inpatient OUD detox than white patients.

The finding that Black and Hispanic patients do not face increased barriers to inpatient detox placement for patients with OUD in our insti- tution is different from previous literature, which found that Black patients have lower odds of obtaining drug and opioid-specific treat- ment [9,10,19] [11,20]. This may be attributed to the makeup of Project ASSERT’s peer support, navigation, network of robust treatment and support services, as well as the provision of transportation. The mem- bers of the Project ASSERT team reflect the diversity of the patients they care for, and a key factor to having minority patients being treated equitably is employing a diverse workforce. Increasing representation, specifically of Black providers, has been identified as a mechanism for dismantling Structural racism [21]. Having a diverse staff has been rec- ognized as a cornerstone for improving treatment for substance use disorder and mental health services nationally [22].

The multivariable logistic GLMM highlighted several associations worthy of further investigation based on whether a barrier to placement occurred. First, the significant interaction between race and sex suggests that women and men may have differences with regard to how their race affects barriers to placement. In our own investigation, only 23

Barrier, n, % per group All visits

(n = 486)

No beds available	345	70.99	137	68.50	208	72.73	0.019
Other	102	20.99	42	21.00	60	20.98	0.413
Did not have meds	53	10.91	23	11.50	30	10.49	0.782
Just left detox	16	3.29	8	4.00	8	2.80	0.803
Barred	8	1.65	7	3.50	1	0.35	0.027
No or inadequate insurance	5	1.03	3	1.50	2	0.70	0.664
No ID or green card	1	0.21	0	0.00	1	0.35	>0.999
No insurance approval	1	0.21	0	0.00	1	0.35	>0.999
Total barriers listed+	531		220		311

* Fisher’s Exact Test.

Black and

Hispanic visits

(n = 200)

White visits

(n = 286)

p-value?

beds are potentially available for women, compared to 117 male beds at the closest and most frequently utilized facilities. Worldwide, there is a dearth of specialized facilities for women to seek OUD treatment [23], despite the fact that women with OUD are more likely than men to have concurrent physical and psychological health problems [24]. In order to comment further on bed availability by sex, a separate targeted investigation would need to be performed to evaluate this.

In large metropolitan areas, it has been found that only 28% of clients complete residential and outpatient OUD treatment, with Black and His- panic patients less likely to complete treatment compared to white pa- tients [25]. Our finding that Black and Hispanic participants were more likely to present multiple times for connection to inpatient detox for

⁺ More than one barrier was listed for some patients.

OUD than white participants has many possible explanations. It may

be that Black and Hispanic patients are more comfortable seeking care after relapse, which could be attributed to positive interactions with our diverse staff at Project ASSERT. However, it more likely is a conse- quence of greater inequities that exist for Black and Hispanic patients, even after acceptance to detox, that predispose this population to re- lapse, and therefore a greater need for re-presentation. Prior literature suggests socioeconomic disparities, including housing and financial in- stability, contribute to lower treatment completion [26]. Sources of bar- riers to housing, employment, and education include the effects of redlining by the Federal Housing Administration in the 1930s, which hindered the accumulation of wealth of Black Americans and contrib- uted to the lack of investment in Black neighborhoods [27,28]. This fur- ther entrenched the war on drugs, which began in the 1970s and fueled the prison industrial complex that disproportionately affected commu- nities of color [29,30]. While there is no statistically significant differ- ence in rates of illicit drug use by race, nearly 80% of people in federal prison and almost 60% of people in state prison for drug offenses are Black and Hispanic [31,32]. It is difficult for people convicted of felony offenses to subsequently obtain gainful employment or access govern- ment assistance such as public housing and financial aid for higher edu- cation, thereby perpetuating the cycle of poverty for people of color that may have initially contributed to their drug use [29].

There are many potential explanations for why racially minoritized

and marginalized individuals may need repeated attempts to success- fully engage with OUD treatment, in spite of no formal barriers identi- fied. In regards to the opioid epidemic, Black and Hispanic intravenous opioid users have been represented in the media in a more menacing and less sympathetic light compared to white counterparts, fueling dis- parate criminalization and stigma [33]. The American Society of Addic- tion Medicine recently acknowledged that drug policy has fueled systemic racism as it relates to addiction treatment, specifically through the criminalization of people of color suffering from substance use dis- orders, and inequitable treatment expansion [34]. Specifically, Black pa- tients have been found to be 35 times less likely than white patients to be prescribed Buprenorphine, an office-based medication to treat OUD, while Black and Hispanic patients are more likely to receive methadone, a highly regulated medication for OUD [34,35]. The regulated dispens- ing practices of methadone have posed as a barrier to employment among those enrolled by limiting time flexibility of job searches and work initiation [36]. Other factors play a role in relapse and re- presentation for OUD treatment, including initiation of MOUD and co- morbid psychiatric conditions; these factors were not analyzed in this study but warrant consideration [37-39].

Time of day, specifically referral after 8 pm, had the strongest associ- ation with a barrier to placement to inpatient detox in both models. We hypothesize that this difference during the 8 pm to 11 pm hours is due in part to beds being already filled and thus unavailable later in the day, or possible decreased staffing at detox facilities in the evening with less capacity to accept patients. Given detox facilities require patients to ar- rive with their outpatient prescriptions, the hospital pharmacy being closed after 8 pm likely also poses a barrier. Overnight, Project ASSERT is closed, so patients seeking referral during those hours would have to come back the next day to be supported in detox placement by a LADC. Improving ways to accept patients into programs at Off hours could reduce barriers to treatment.

Lastly, male patients with Medicaid insurance, who made up the ma- jority of participants in the study, were less likely to face a barrier to placement. Prior studies found that while Medicaid expansions under the Affordable Care Act led to an increase in outpatient OUD treatment, utilization of inpatient detox was unchanged [40]. In the US, only 5.9% of SUD facilities provide treatment in an inpatient setting, whereas 81.4% provide outpatient treatment [41]. It is possible that our findings of fewer barriers for Medicaid patients can be explained by the higher proportion of inpatient detox facilities in this state that accept state- based coverage and Medicaid. While findings are mixed and out- patient MOUD treatment has become a mainstay, for certain patient

populations, an inpatient treatment setting may be more beneficial for some patient populations suffering with OUD, as they remove social triggers for relapse [42-44].

Although it was not a study objective, it is important to note that the type of barrier encountered was different based on race. Overall, staff were most likely to list bed availability as the major barrier to place- ment; however white participants encountered issues with bed avail- ability more frequently. This likely accounts for why white men more often listed a barrier to placement. Further investigation into associated factors for white men such as arrival time, facility preference and insur- ance status should be explored as reasons for limited bed availability in this population.

An important area to not ignore was that Black and Hispanic partic- ipants were more likely to be barred from treatment facilities than white participants (3.50% vs 0.35%, p = 0.027). Being barred is due to a detox center’s personnel determinations, based on record of past be- havior. It is possible that this reflects racial injustice among patients being barred from treatment. One potential mechanism is through the role of interpersonal racism due to preconceived notions of hostility and aggression, particularly with regards to the stigmatization of Black men [45]. This finding warrants further investigation among broader geographic areas to determine the association of race/ethnicity on being barred from treatment for OUD.

• 1. Limitations

Retrospective Cohort study design was necessary to achieve the aims of this study. However the data in this study was collected for clinical purposes, and the accuracy of data was limited to that collected as part of routine clinical care. Our data were dependent on the LADC, registra- tion, and medical staff documentation, and it is unclear how irregularities in documentation would have affected our results. For example, “no bed available” is the first option in the dropdown menu for a barrier to detox; and therefore, could have been selected erroneously for ease.

We did not evaluate for racial equity in referrals to Project ASSERT from the ED. Due to inconsistent documentation, we were unable to ad- equately evaluate for inequities in this process. We therefore began our study after the patient was registered at Project ASSERT to evaluate for barriers to care when documentation is standardized. Since this investi- gation, we have begun to explore how to standardize documentation of referrals to Project ASSERT from the ED.

While the participants included in the study did present to both the ED and Project ASSERT during the study period, they did not necessarily present to Project ASSERT immediately following an ED visit. We chose this approach as patients may not be ready to enter an inpatient detox at time of ED visit however may opt to return later, as Project ASSERT staff are available for walk-in visits during all open hours. Additionally, the period that we assessed was longer for detox referrals than ED visits. We did this to allow for delayed visits after the ED visit if the patient sought detox later when personally ready. Since there are multiple other hospitals in Boston, we are unable to ensure participants did not seek care or referral through a different healthcare system. It is un- known if this could have affected our results.

For 21% of encounters with a barrier to placement, a barrier of “other” was listed with no further specification. We were unable to identify the types of barriers that existed for those participants via our dataset, however a prior qualitative study among detox program pro- viders identified other barriers to access detox. These included program rules, admission requirements, and funding shortages; patient-centered causes included stigma related to a SUD diagnosis or receiving detox services, a deficiency of education and training, personal motivations and competing responsibilities [46].

While we found that Black and Hispanic individuals were more likely to have 3 or more visits seeking OUD detox than white people, we did not extract from the data whether the patients met a barrier at each previous visit, leaving the possibility that patients experienced

the same barrier multiple times. For this reason, further work needs to be done to evaluate the specific barriers among patients with numerous visits for linkage to inpatient detox.

We chose to select participants based on OUD as their primary substance for which they sought detox. We did not adjust for multi- ple types of substances (e.g. concurrent cocaine or alcohol use), which in some cases may have made it more challenging to place patients at a facility.

We also recognize the limitations of inpatient detox as a component of OUD treatment, both with regard to cost and overall efficacy [44,47]. Nevertheless, we chose to study barriers to inpatient detox placement as this is the treatment option most of our patients select. It is unclear the rationale for this, however it may be due to several factors such as feelings of needing a different environment to be successful, the support of treatment facility staff, or availability of comfort medications. While all of the inpatient detox facilities to which we refer patients offer MOUD, we also were unable to assess if patients transition to mainte- nance therapy after discharge from detox, which may have affected our findings particularly with regard to repeated ED visits for OUD detox placement.

• 1. Implications for practice

Our results support that despite historic lower treatment rates for Black and Hispanic patients, in the context of near-universal health insurance coverage, inequities in barriers to being accepted to inpatient OUD detox facilities can be mitigated. The findings of this study are critical in informing hospital-based programs to connect patients to OUD treatment. Given that the ED is a large touchpoint for healthcare for many of these patients, ED-based programs to increase accessibility to treatment is important [48]. Next, the programs themselves must be created using a community-partnered approach, represent the racial and ethnic diversity of the communities that they serve, and understand the intricacies and lived experiences of those with substance use disor- der [14]. These qualities will encourage robust rapport-building with patients, which is likely what contributed to the findings of this study. Lastly, the implementation of regulations and resource allocation for treatment facilities to accept Medicaid patients is critical for access to care in this population.

1. Conclusion

We did not find that Black and Hispanic patients faced increased bar- riers for inpatient detox placement. We did find among men however, those who were Black or Hispanic had decreased odds of having a bar- rier compared to those who were white. We believe this unexpected result is due in part to our institution’s community-based, diverse team of LADCs who focus on relationship-building with patients and provide comprehensive care, as well as state regulations supporting Medicaid admissions at inpatient OUD detox facilities. Additionally, Black and Hispanic patients were more likely to present multiple times to the ED seeking connection to inpatient OUD detox than white patients. This suggests Black and Hispanic patients may be more com- fortable seeking care with our team after relapse, or-more likely-that these patients experience inequities once they leave the ED, leading to higher relapse rates. Further work is needed to elucidate and implement efforts that result in mitigating barriers for Black and Hispanic patients, but also determine why Black and Hispanic patients have more repeat

acquisition and analysis. All authors were responsible for interpretation of the data and drafting the manuscript.

Credit authorship contribution statement

Jessica Faiz: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Ed Bernstein: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Julianne N. Dugas: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Elissa M. Schechter-Perkins: Conceptualization, Formal analysis, Inves- tigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Lauren Nentwich: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Kerrie P. Nelson: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Emily C. Cleveland Manchanda: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Ludy Young: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing. Joseph R. Pare: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing - review & editing.

Declaration of Competing Interest

No authors have any relevant conflict of interests to disclose.

Acknowledgments

The Project ASSERT staff who work tirelessly to support patients with OUD. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Appendix A. ICD-10 code list

Diagnosis ICD10

Opioid abuse, uncomplicated F11.10

Opioid abuse, in remission F11.11

Opioid abuse with intoxication, uncomplicated F11.120

Opioid abuse with intoxication delirium F11.121

Opioid abuse with intoxication, unspecified F11.129

Opioid abuse with opioid-induced mood disorder F11.14

Opioid abuse with unspecified opioid-induced disorder F11.19

Opioid dependence, uncomplicated F11.20

Opioid dependence, in remission. F11.21

Opioid dependence with intoxication, uncomplicated F11.220

Opioid dependence with intoxication delirium F11.221 Opioid dependence with intoxication with perceptual disturbance F11.222 Opioid dependence with intoxication, unspecified F11.229

Opioid dependence with withdrawal F11.23

Opioid dependence with opioid-induced mood disorder F11.24 Opioid dependence with opioid-induced psychotic disorder, unspecified F11.259 Opioid dependence with unspecified opioid-induced disorder F11.29 Opioid use, unspecified, uncomplicated F11.90

Opioid use, unspecified with intoxication, uncomplicated F11.920

Opioid use, unspecified with intoxication, unspecified F11.929

Opioid use, unspecified with withdrawal F11.93

Opioid use, unspecified with opioid-induced mood disorder F11.94

ED visits for referral to OUD treatment.

Opioid use, unspecified with opioid-induced psychotic disorder, unspecified.

F11.959

Author contributions

JF, EB, JND, LN, KPN, ESP, LY and JRP conceived the study and design. LY provided institutional knowledge and operational background for Project ASSERT. JF, EB, JND, KPN, and JRP were responsible for data

Opioid use, unspecified with unspecified opioid-induced disorder F11.99 Poisoning by heroin, accidental (unintentional), initial encounter T40.1X1A Poisoning by heroin, intentional self-harm, initial encounter T40.1X2A Poisoning by heroin, undetermined, initial encounter T40.1X4A Poisoning by other opioids, accidental (unintentional), initial encounter T40.2X1A Poisoning by other opioids, accidental (unintentional), subsequent T40.2X1D

(continued)

Diagnosis ICD10

encounter

Poisoning by other opioids, intentional self-harm, initial encounter T40.2X2A Poisoning by other opioids, undetermined, initial encounter T40.2X4A Adverse effect of other opioids, initial encounter T40.2X5A

Underdosing of other opioids, initial encounter T40.2X6A Poisoning by methadone, accidental (unintentional), initial encounter T40.3X1A Adverse effect of methadone, initial encounter T40.3X5A

Underdosing of methadone, initial encounter T40.3X6A

1. Bailey ZD, Feldman JM, Bassett MT. How structural racism works — racist policies as a root cause of U.S. racial health inequities. N Engl J Med. 2020. https://doi.org/10. 1056/nejmms2025396.
2. Workforce development needs in the field of substance use disorders a Report from Department of Alcohol and Drug Programs; 2013.
3. INTERVIEW: Head of UN drug body urges greater access to treatment for women; 2017.
4. Koons AL, Greenberg MR, Cannon RD, Beauchamp GA. Women and the experience of pain and opioid use disorder: a literature-based commentary. Clin Ther. 2018; 40:190-6. https://doi.org/10.1016/j.clinthera.2017.12.016.
5. Stahler GJ, Mennis J. Treatment outcome disparities for opioid users: are there racial and ethnic di ff erences in treatment completion across large US metropolitan areas

Poisoning by other synthetic narcotics, accidental (unintentional), initial encounter

Poisoning by other synthetic narcotics, intentional self-harm, initial

encounter

T40.4X1A

T40.4X2A

? Drug Alcohol Depend. 2018;190:170-8. https://doi.org/10.1016/j.drugalcdep. 2018.06.006.

1. Saloner B, Cook BL. Blacks and hispanics are less likely than whites to complete ad- diction treatment, largely due to socioeconomic factors. Health Aff. 2013;32:135-45.

Poisoning by other synthetic narcotics, undetermined, initial encounter T40.4X4A Adverse effect of other synthetic narcotics, initial encounter. T40.4X5A

https://doi.org/10.1377/hlthaff.2011.0983.

1. Heard-Garris N, Boyd R, Kan K, Perez-Cardona L, Heard NJ, Johnson TJ. Structuring poverty: how racism shapes child poverty and child and adolescent health. Acad

Poisoning by, adverse effect of and underdosing of narcotics and

psychodysleptics [hallucinogens]

Poisoning by unspecified drugs, medicaments and biological substances, accidental (unintentional), initial encounter

References

T40.4X6A

T50.901A

Pediatr. 2021;21:S108-16. https://doi.org/10.1016/j.acap.2021.05.026.

1. Rothstein R. The color of law: A forgotten history of how our government segregated

America; 2017.

1. Moore LD, Elkavich A. Who’s using and who’s doing time: incarceration, the war on drugs, and public health. Am J Public Health. 2008;98:782-6. https://doi.org/10. 2105/AJPH.2007.126284.
2. Roberts Dorothy E. The social and moral cost of mass incarceration in African American communities. Stanford Law Rev. 2004;56:1271-305.
3. Dydyk AM, Jain NK. GMOUD. Opioid Use Disorder. StatPearls; 2020. https://www- ncbi-nlm-nih-gov.ezproxy.bu.edu/books/NBK553166/.
4. Volkow N. Opioid overdose crisis. https://www.drugabuse.gov/drugs-abuse/opioids/

opioid-overdose-crisis; 2020. (accessed July 14, 2020).

1. Strayer RJ, Hawk K, Hayes BD, Herring AA, Ketcham E, Lapietra AM. Management of opioid use disorder in the emergency department: A white paper prepared for the American academy of emergency medicine; 2019.
2. Larochelle MR, Bernstein R, Bernson D, Land T, Stopka TJ, Rose AJ, et al. Touchpoints - opportunities to predict and prevent opioid overdose: a cohort study. Drug Alcohol Depend. 2019;204:107537. https://doi.org/10.1016/j. drugalcdep.2019.06.039.
3. White C. Progress Against the Opioid Epidemic Is Not Reaching Black Americans. Heal City; 2021.https://www.massheal.org/healthcity/population-health/progress- against-opioid-epidemic-not-reaching-black-americans (accessed March 9, 2021).
4. Massachusetts Department of Public Health. Opioid-Related Overdose Deaths, All In- tents, MA Residents - Demographic Data Highlights; 2019; 26-30.
5. Khatri UG, Pizzicato LN, Viner K, Bobyock E, Sun M, Meisel ZF, et al. Racial/ethnic dis- parities in unintentional fatal and nonfatal emergency medical services-attended opioid overdoses during the COVID-19 pandemic in Philadelphia. JAMA Netw Open. 2021;4:e2034878. https://doi.org/10.1001/jamanetworkopen.2020.34878.
6. Opioid-related overdose deaths soar among Black residents; May 12, 2021.
7. Gryczynski J, Schwartz RP, Salkever DS, Mitchell SG, Jaffe JH. Patterns in admission delays to outpatient methadone treatment in the United States. J Subst Abuse Treat. 2011;41:431-9. https://doi.org/10.1016/j.jsat.2011.06.005.
8. D’Aunno T, Pollack HA. Changes in methadone Treatment practices: results from a national panel study, 1988-2000. JAMA. 2002;288:850-6. https://doi.org/10.1001/ jama.288.7.850.
9. Lagisetty PA, Ross R, Bohnert A, Clay M, Maust D. Buprenorphine treatment divide by race/ethnicity and payment. JAMA Psychiat. 2019;76:979-81. https://doi.org/10. 1097/EDE.0000000000000858.
10. Mehta A, Murphy J, Wada R, Griffith J, Erdman E. DATA BRIEF: racial differences in accessing substance use disorder treatment following an opioid overdose in massa- chusetts and boston; 2015.
11. Jones CPJ. Levels of racism: a theoretic framework and a Gardener’s tale. Am J Public Health. 2020;90:1212-5. https://doi.org/10.3102/0002831211424313.
12. Bernstein E, Bernstein J, Levenson S. l- ‘ roject ASSERT: An ED-based intervention to increase access to primary care, Preventive services, and the substance abuse treat- ment system; 1997.
13. Boyd Rhea W, Lindo Edwin G, Weeks Lachelle D, McLemore Monica R. On racism: a new standard for publishing on racial health inequities | health affairs. Heal Aff Blog. 2020:1-13. https://doi.org/10.1377/hblog20200630.939347.
14. Harris Paul A, Taylor Robert, Thielke Robert, Payne Jonathon, Gonzalez Nathaniel, Conde JG. Research electronic data capture (REDCap) - A metadata-driven method- ology and workflow process for providing translational research informatics sup- port. J Biomed Inform. 2009;42:377-81.
15. Kelly JF, Greene MC, Bergman BG, White WL, Hoeppner BB. How many recovery at- tempts does it take to successfully resolve an alcohol or drug problem? Estimates and correlates from a National Study of recovering U.S. Adults Alcohol Clin Exp Res. 2019;43:1533-44. https://doi.org/10.1111/acer.14067.
16. PASS 2019 power analysis and sample size software; 2019.
17. Wu L, Zhu H, Swartz MS. Treatment utilization among persons with opioid use dis- order in the United States. Drug Alcohol Depend. 2016;169:117-27. https://doi.org/ 10.1016/j.drugalcdep.2016.10.015.
18. Hansen Helena, Siegel Carole, Case Brady, Bertollo David, DiRocco Danae, Galanter M. Variation in use of buprenorphine and methadone treatment by racial, ethnic and in- come characteristics of residential social areas in new York City. J Behav Health Serv Res. 2013;40:1-14. https://doi.org/10.1007/s11414-013-9341-3.Variation.
19. Alliance DP. The drug war, mass incarceration and race. Drug policy alliance; 2016; 1-3.
20. Services USD of H and H. Results from the 2013 National Survey on Drug Use and Health: Summary of National Findings, vol. 131 A; 2014.https://doi.org/10.1002/ ajmg.a.30231.
21. Netherland Julie, Hansen H. The war on drugs that wasn’t: wasted whiteness, “dirty doctors,” and race in media coverage of prescription Opioid misuse. Cult Med Psychi- atry. 2016;40:664-86. https://doi.org/10.1007/s11013-016-9496-5.The.
22. American Society of Addiction Medicine. public policy statement on advancing racial justice in addiction medicine. Public policy statement. 2021;1-6. https://www.asam. org/docs/default-source/public-policy-statements/asam-policy-statement-on- racial-justiced7a33a9472bc604ca5b7ff000030b21a.pdf?sfvrsn=5a1f5ac2_2 [Accessed 9 March 2021].
23. Mann B. drug overdose deaths Surge Among Black Americans During Pandemic. NPR; 2021.https://www.npr.org/2021/03/03/970964576/drug-overdose-deaths- surge-among-black-americans-during-pandemic [Accessed 9 March 2021].
24. Richardson L, Wood E, Montaner J, Kerr T. Addiction treatment-related employment barriers: the impact of methadone maintenance. J Subst Abuse Treat. 2012;43: 276-84. https://doi.org/10.1016/j.jsat.2011.12.008.
25. Ben Abdallah A, Phelps DL, Ph D, Spitznagel EL, Ph D, Horton JC. Substance depen- dence and other psychiatric disorders among drug dependent subjects. Race Gender Correlates. 2000:113-25.
26. Boyd MR, Phillips K, Dorsey CJ. Alcohol and other drug disorders, comorbidity, and violence: comparison of rural African American and Caucasian women. Arch Psychiatr Nurs. 2003;17:249-58. https://doi.org/10.1053/j.apnu.2003.10.003.
27. Tkacz J, Severt J, Cacciola J, Ruetsch C. Compliance with buprenorphine medication- assisted treatment and relapse to opioid use; 2011; 55-62. https://doi.org/10.1111/j. 1521-0391.2011.00186.x.
28. Meinhofer A, Witman AE. The role of health insurance on treatment for opioid use disorders: Evidence from the Affordable Care Act Medicaid expansion ?, 60; 2018; 177-97.
29. Cummings Janet R, Wen Hefei, Ko Michelle, Druss BG. Race/ethnicity and geographic access to medicaid substance use disorder treatment facilities in the United States, 71; 2014; 190-6. https://doi.org/10.1001/jamapsychiatry.2013.3575.
30. Stahler GJ, Mennis J, DuCette JP. Residential and outpatient treatment completion for substance use disorders in the U.S.: moderation analysis by demographics and drug of choice. Addict Behav. 2016;58:129-35. https://doi.org/10.1016/j.addbeh.2016.02.030.
31. Day E, Strang J. Outpatient versus inpatient opioid detoxification: a randomized con- trolled trial. J Subst Abuse Treat. 2011;40:56-66. https://doi.org/10.1016/j.jsat.2010. 08.007.
32. Wakeman SE, Larochelle MR, Ameli O, Chaisson CE, McPheeters JT, Crown WH, et al. Comparative effectiveness of different treatment pathways for opioid use disorder. JAMA Netw Open. 2020;3:1-12. https://doi.org/10.1001/jamanetworkopen.2019. 20622.
33. Simon KM. Them and Me — the care and treatment of black boys in America. N Engl J

Med. 2020;383:1904-5. https://doi.org/10.1056/nejmp2022606.

1. Schultz NR, Martinez R, Cucciare MA, Timko C. Patient, program, and system barriers and facilitators to detoxification services in the U.S. Veterans Health Administration: a qualitative study of provider perspectives. Subst Use Misuse. 2016;51:1330-41. https://doi.org/10.3109/10826084.2016.1168446.
2. Mojtabai R, Zivin JG. Effectiveness and cost-effectiveness of four treatment modali- ties for substance disorders: a Propensity score analysis. Health Serv Res. 2003;38: 233-59. https://doi.org/10.1111/1475-6773.00114.
3. Khatri UG, Samuels EA, Xiong R, Marshall BDL, Perrone J, Delgado MK. Variation in emergency department visit rates for opioid use disorder: implications for quality improvement initiatives. Am J Emerg Med. 2022;51:331-7. https://doi.org/10. 1016/j.ajem.2021.10.047.
4. Fisher RA. Statistical methods for research workers. 5th. Edinburgh: Oliver & Boyd; 1934.