Expanding buprenorphine prescribing in primary care: a qualitative study of the experiences of primary care providers and nurse care managers participating in the New York City buprenorphine nurse care manager initiative

Abstract

Background

Improving access to and retention in evidence-based treatment for opioid use disorder, including buprenorphine, is a critical response to the opioid overdose crisis. To increase the availability of buprenorphine treatment in primary care settings, the New York City Department of Health and Mental Hygiene implemented the Buprenorphine Nurse Care Manager Initiative in safety-net primary care clinics. The initiative funds nurse care managers to coordinate ongoing buprenorphine care and provides clinical support and technical assistance for implementation.

Methods

As part of a process evaluation of the initiative, we conducted in-depth interviews with 18 primary care providers new to prescribing buprenorphine and five nurse care managers across six organizations in New York City that participated in the Buprenorphine Nurse Care Manager Initiative between 2017 and 2019. We aimed to understand participating providers’ views on the successes and challenges of the initiative. Thematic and trajectory analytic approaches were used to capture major themes and changes over time.

Results

Findings show that participating providers valued many aspects of the initiative, suggesting that integrating buprenorphine treatment with the support of a nurse care manager into safety-net primary care clinics can effectively expand access to buprenorphine and quality of care for people with opioid use disorder in New York City.

Conclusions

Findings from this process evaluation can inform future primary care-based buprenorphine treatment initiatives. Recommendations include ensuring ample nurse care manager support for primary care providers, robust mentorship structures, and organizational buy-in for initiative sustainability.

Background

Opioid overdose is a public health crisis in the United States. Nationally, the number of drug overdose deaths more than quadrupled from 2000 to 2019, and opioid overdose was declared a nationwide public health emergency in 2017 [1]. In New York City (NYC), the rate of overdose death in 2023 remained stable from 2022, when the rate was the highest since reporting began in 2000 at 44.7 per 100,000. In 2023, 83% of all overdose deaths in NYC involved an opioid [2, 3]. There are wide disparities in overdose death by race and ethnicity, age, income, and neighborhood of residence, reflecting underlying inequities in socioeconomic status, employment opportunities, educational attainment, involvement with the criminal legal system, and access to housing that stem from long-standing racism and community disinvestment. Black and Hispanic/Latino/a New Yorkers, and residents of very high poverty neighborhoods in NYC continued to bear a disproportionate burden of overdose deaths in 2023 [3].

Addressing the overdose crisis demands a firm commitment to harm reduction, an approach that encompasses a broad spectrum of strategies aimed at mitigating both the conditions and consequences of drug use. Practices such as overdose prevention, safer drug use education, wound care, and treatment are delivered within a framework grounded in the principles of person-centered, nonjudgmental, and nonpunitive care. A harm reduction approach respects an individual’s autonomy to set their own goals for health and well-being and recognizes that abstinence is not the desired outcome for everyone.

Expanding access to evidence-based treatment for opioid use disorder (OUD), particularly to populations experiencing disproportionately high rates of overdose, is critical to addressing opioid-related morbidity and mortality. Buprenorphine and methadone are highly effective medication treatments for OUD,1 reducing drug use and death and improving social well-being and functioning [4, 5]. Buprenorphine and methadone work by reducing or eliminating withdrawal symptoms, blocking the effects of other opioids, and reducing cravings to use opioids. While methadone has the longest history of all pharmacological treatments for OUD, federal regulations limit its provision to observed daily dosing at highly regulated opioid treatment programs, with the option of take-home doses only after program staff deems the person “adherent.” Buprenorphine treatment, by contrast, offers increased flexibility and carries less stigma because it can be prescribed in office-based and general practice settings and taken at home like any other medication [6]. While buprenorphine was approved for OUD treatment by the Food and Drug Administration in 2002, the regulatory framework for its use in office-based opioid treatment (OBOT) was established through the Drug Addiction Treatment Act in 2000. Offering OBOT in primary care settings is an effective harm reduction approach to providing care and improving health outcomes among people who use drugs.

The breadth of settings in which buprenorphine can be offered should result in widespread access to the medication, yet a variety of circumstances have limited its availability. The average medical school curriculum dedicates only a few hours to training on addiction, and only 56% of residency programs require any training in substance use disorders [7, 9]. Most health professionals are not sufficiently trained to educate patients about drug use and addiction, conduct screening and interventions for risky drug use, or diagnose and treat substance use disorders [10]. Moreover, until 2020, prescribing buprenorphine to treat OUD necessitated physicians take a federally regulated eight-hour training course, increasing to 24-h of training for nurse practitioners and physician assistants, and receive authorization (known as a “waiver”) from the Drug Enforcement Administration.2 These factors have contributed to a cultural sensibility among providers, practice managers, and administrators that substance use treatment, including buprenorphine prescribing, is outside the scope of primary care.

In addition to limited exposure to treating substance use disorders in general and a lack of training to prescribe buprenorphine in particular, clinicians cite a variety of barriers to offering buprenorphine in their practices. Stigma among health professionals toward people who use drugs is pervasive [11]. In addition, clinicians face myriad barriers, including insufficient nursing and ancillary support, lack of time to and interest in managing what is presumed to be a difficult patient population, and lack of institutional backing [12–14].

Although the availability of buprenorphine treatment has increased nationally over the last two decades, growth in buprenorphine treatment access has been disproportionally greater in urban counties and in ZIP codes with higher incomes and a higher percentage of White residents, compared to that of rural counties and ZIP codes with lower incomes and a higher percentage of Black and Hispanic/Latino/a residents, respectively [15, 16]. National buprenorphine utilization data follows the same pattern: from 2004 to 2015, clinical visits for buprenorphine grew fastest among White patients and patients with private insurance, compared to Black patients and patients with public insurance [17]. A 2013 study in NYC showed that buprenorphine treatment rates were highest in ZIP codes with the highest incomes and lowest among neighborhoods with predominantly Black and Hispanic/Latino/a residents, whereas methadone treatment rates were highest in ZIP codes with the highest percentages of low income and Hispanic/Latino/a residents [18]. Researchers continue to report disparities in buprenorphine treatment rates across ZIP codes in NYC, advocating for expanded buprenorphine prescribing within safety-net health care settings that offer care to people regardless of their ability to pay or insurance status and primarily service neighborhoods bearing the highest burden of overdose deaths [19].

As part of a multi-pronged response to the overdose crisis and the limited options for buprenorphine treatment in settings that serve low-income and racialized marginalized populations, the NYC Department of Health and Mental Hygiene (NYC Health Department) launched the Buprenorphine Nurse Care Manager Initiative (BNCMI) in 2016 [20]. Rooted in harm reduction principles, the goal of BNCMI was to increase the availability and provision of buprenorphine treatment for OUD via OBOT in NYC safety-net primary care clinics. BNCMI provided funding for a registered nurse as a nurse care manager (NCM) to manage and support the practice offering buprenorphine treatment, provided technical assistance for implementation, and offered clinical mentorship to primary care providers (PCPs)3 newly prescribing buprenorphine from buprenorphine experts around the city. Fifteen clinics operated by six agencies were awarded funds to implement BNCMI in 2016 with an additional twelve clinics operated by eight separate agencies funded in 2018.

This paper explores the experiences of NCMs and PCPs prescribing buprenorphine during the first year of BNCMI implementation and offers recommendations to advance office-based buprenorphine as a means of expanding access to treatment and reducing opioid-related morbidity and mortality among underserved populations. We use Rogers’ Diffusion of Innovation Theory to interpret the results of the study and frame the discussion of successes and challenges of BNCMI as it was implemented over the course of a year at the six primary care agencies that participated in this study [21]. The Diffusion of Innovation Theory has been used, adapted, and expanded on in a wide variety of health settings to identify facilitators and barriers to implementing and sustaining programs, and more generally to support public health program planning and evaluation [22]. The theory has been utilized as a framework in many qualitative studies in myriad research areas such as developing primary care models for managing dementia, scaling up antiviral therapy interventions for hepatitis C virus infection, and assessing perceptions of contraception methods [23–25]. In the substance use field, examples include using Rogers' theory to explore the adoption and implementation of computer-assisted therapy for substance use, an opioid overdose prevention program, and buprenorphine treatment within integrated health plans [26–28]. While multiple adaptations and expansions of the theory have been developed, in this study we use theoretical concepts as originally envisioned and defined by Rogers [21].

The core components of BNCMI are examined together as an “innovation” through the lens of the Diffusion of Innovation Theory. Several key concepts from Rogers’ theory are defined and used to frame the discussion. Previous research on the NYC BNCMI and similar primary care-based buprenorphine treatment initiatives has been focused on implementation and patient characteristics [20, 29], patient experience [30], retention and attrition [31, 32], and financing 33. To our knowledge, this is the first qualitative study to examine the views of NCMs as well as of PCPs new to prescribing buprenorphine and supported by the NCM initiative. Furthermore, we explored how the study participants’ perspectives on prescribing buprenorphine evolved over the course of the first year of BNCMI implementation. While previous research has measured BNCMI’s success in reaching marginalized patient populations and improving patient outcomes, this paper offers key insights into how best to support clinicians who are newly undertaking this work.

Methods

Study setting

The NYC BNCMI was implemented in 27 clinics within 14 safety-net primary care agencies citywide, including at least one clinic within each of the 5 boroughs of NYC. As part of BNCMI, each participating agency received funding for one full-time NCM. Agencies often assigned the NCM to more than one of their clinics. Each clinic had at least four PCPs with waivers to prescribe buprenorphine for the treatment of opioid use disorder, including physicians (MDs/DOs), nurse practitioners (NPs), and physician assistants (PAs).4 NYC Health Department provided individualized clinical mentorship from buprenorphine experts and hosted quarterly Learning Community meetings for PCPs and NCMs. Learning Community meetings focused on subjects including home initiation, harm reduction, and racial equity [30]. Leadership involvement in the initiative varied by clinic site.

Procedures

The six agencies represented in the study comprised a total of 15 clinics (multiple clinics per agency). Waivered PCPs and NCMs from all clinics were invited by NYC Health Department research staff to participate in an in-depth interview at two time points as part of the process evaluation. PCPs were contacted for their first interview within the first six months of the initiative, either when they had prescribed buprenorphine to at least three patients or for at least three months (whichever came first) to capture early (Time 1) experiences with BNCMI. When the first PCP reached eligibility at an agency, the research staff was notified by the BNCMI staff implementing the initiative. The research staff then contacted the NCM for their first interview. One year after their first interview, NCMs and PCPs were interviewed a second time to capture changes in practices, perspectives, or beliefs (Time 2).

Between July 2017 and November 2019, in-depth, semi-structured interviews at both timepoints were conducted with a total of 18 PCPs and five NCMs. Only five of the six NCM interviews were included in the analysis as one NCM left the position prior to the second interview. All procedures were reviewed by the NYC Health Department’s Institutional Review Board. NYC Health Department research staff conducted data collection and analysis independent of the NYC Health Department BNCMI implementation team.

Data collection

Interviews were semi-structured and conducted in person with PCPs and NCMs at their place of work (one interview took place at NYC Health Department). For PCPs, the interview topics included prior experience working with people who use drugs, current buprenorphine patient panel and prescribing practices (including around treatment initiation), relationship with patients and NCMs, lessons learned from participating in the initiative, and successes and challenges of the initiative. For NCMs, the interview topics included prior experience working with people who use drugs, initiative workflows and protocols, strategies for patient engagement and retention, buprenorphine treatment practices, approaches to care, relationships with patients and PCPs, and successes and challenges of the initiative. Basic self-reported demographic data were collected at the beginning of the interview. All interviews were audio-recorded and professionally transcribed for analysis. The mean duration of the interviews was approximately 41 min (range 19–72).

Analysis

Descriptive statistics for the sample were calculated using SPSS (IBM SPSS Statistics for Windows, 2017) and included demographic data and other basic participant characteristics (race and ethnicity, gender, age, years of practice, and credentials). Gender and race and ethnicity categories are reported as collected in 2018 for data accuracy. We recognize that some language used does not reflect current best practices in reporting demographic characteristics. The categories "Transgender (MTF)" and "Transgender (FTM)" (for which participant responses were both n = 0) were collapsed into a larger category, "Transgender".

Transcripts were entered into Dedoose (“Dedoose Version 8.0.35, web application for managing, analyzing, and presenting qualitative and mixed method research data,” 2018) for data management and creating a qualitative codebook. Research staff developed the codebook by incorporating both a priori codes (based on study goals) and emergent codes.

To capture key themes related to the experiences of PCPs and NCMs at both time points, changes and continuities over time, and agreements and disagreements among study participants, thematic and trajectory analytic approaches were utilized [34–36]. Methodological rigor was enhanced by maintaining an audit trail documenting analytic decisions and emerging themes, applying analytic memos when coding, and periodic debriefing with other members of the research team [37]. Analytic rigor was also ensured by double coding transcripts; 98% (n = 45) of transcripts were coded independently by two members of the research team and subsequently reconciled to establish coding reliability.

A series of tables, referred to as matrices in this study, was used to organize, display, and systematically analyze the data, allowing for identification of patterns and themes across different dimensions. First, all excerpts that had been assigned the codes “Successes of the NCM model” or “Challenges of the NCM model” were displayed in Matrix 1, by agency, interviewee, and time point. Matrix 2 examined themes grouped into categories across participant interviews at both times within a single agency. Changes and continuities over time and agreements and disagreements among participants within a single agency were also recorded, along with major themes for each category. Finally, major themes that appeared at any point in the data, noteworthy changes and continuities over time, and important agreements and disagreements by category from Matrix 2 were inputted into Matrix 3 for each agency; responses were then compared across agencies by category. Suggestions for program improvement made by study participants at any time point were also pulled from Matrix 3. The Diffusion of Innovation Theory was used as an interpretative framework to further examine the findings from the study in the Discussion section (Rogers, 1962).

Researcher characteristics

In total, three researchers (SS, NB, and ECW) participated in various steps of data collection and analysis for this study. SS and NB conducted the interviews, all three coded sections of the transcripts, and NB and ECW developed the series of matrices and conducted the analysis. All three researchers were NYC Health Department staff members at the time, with backgrounds in substance use and qualitative research. None was involved in BNCMI implementation, operations, or ongoing monitoring.

The researchers’ harm reduction approach to substance use care might have influenced how they interpreted providers’ perspectives and analyzed themes related to approaches to care or comfort with prescribing buprenorphine.

Results

Participant characteristics

As seen in Table 1, in-depth interviews with a total of 18 buprenorphine PCPs and five NCM participants from six of the BNCMI agencies were included as part of this analysis. The total sample of 23 participants predominantly identified as female (73.9%) with a mean age of 45.0 years. Most NCMs self-identified as Hispanic/Latino/a (60.0%) and two NCMs identified as White non-Hispanic/Latino/a (40.0%). Among PCPs, 33.3% identified as White non-Hispanic/Latino/a, 27.7% as Black non-Hispanic/Latino/a, 22.2% as Asian, and 16.7% as multiracial non-Hispanic/Latino/a. All NCMs were registered nurses (RN). Most PCPs were medical doctors (MD) (61.1%), and the remainder were nurse practitioners (NP) (27.8%) and physician assistants (PA) (11.1%). All NCMs had prior experience working with people who use drugs and 38.9% of PCPs had prior training in addiction medicine (aside from medical school or the buprenorphine waiver training). Among all study participants the average number of years practicing medicine was 10.2 years, ranging widely from one year to 36 years. Please see Table 1 for more demographic data and other participant characteristics.

Table 1.

Overview of study participant characteristics

Variable	Primary care providers	Nurse care managers	Total
Total sample size	N = 18	N = 5	N = 23
Gender	% (n)	% (n)	% (n)
Female	72.2 (13)	80.0 (4)	73.9 (17)
Male	27.8 (5)	20.0 (1)	26.1 (6)
Transgender	0.0 (0)	0.0 (0)	0.0 (0)
Other	0.0 (0)	0.0 (0)	0.0 (0)
Age	Years	Years	Years
Mean, standard deviation	44.4, 11.8	47.0, 6.2	45.0, 10.9
Median, range	39.5, 28–63	51.0, 38–53	41.0, 28–63
Race and ethnicity	% (n)	% (n)	% (n)
Black/African American/African	27.7 (5)	0.0 (0)	27.1 (5)
White/Caucasian	33.3 (6)	40.0 (2)	34.8 (8)
Hispanic/Latino/a	0.0 (0)	60.0 (3)	13.0 (3)
Asian	22.2 (4)	0.0 (0)	17.4 (4)
Multi-racial	16.7 (3)	0.0 (0)	13.0 (3)
Native Hawaiian/Pacific Islander	0.0 (0)	0.0 (0)	0.0 (0)
Native American	0.0 (0)	0.0 (0)	0.0 (0)
Other	0.0 (0)	0.0 (0)	0.0 (0)
Credentials	% (n)	% (n)	% (n)
Physician (MD/DO)	61.1 (11)	N/A	47.8 (11)
Physician assistant (PA)	11.1 (2)	N/A	8.7 (2)
Nurse practitioner (NP)	27.8 (5)	N/A	21.7 (5)
Registered nurse (RN)	N/A	100.0 (5)	21.7 (5)
Prior training in addiction medicine (PCPs) or prior experience working with people who use drugs (NCMs) (% Yes)	% (n)	% (n)	% (n)
	38.9 (7)	100.0 (5)	52.2 (12)
Number of years practicing	Years	Years	Years
Mean, standard deviation	10.7, 10.2	8.1, 7.3	10.2, 9.7
Median, range	8.0, 1–36	6.0, 1.5–22	6.0, 1–36
Agency	% (n)	% (n)	% (n)
Agency 1	27.8 (5)	0.0 (0)	21.7 (5)
Agency 2	11.1 (2)	20.0 (1)	13.0 (3)
Agency 3	16.7 (3)	20.0 (1)	17.4 (4)
Agency 4	16.7 (3)	20.0 (1)	17.4 (4)
Agency 5	11.1 (2)	20.0 (1)	13.0 (3)
Agency 6	16.7 (3)	20.0 (1)	17.4 (4)

Unless indicated otherwise in the methods section, gender and race categories are presented as collected in 2018 for data accuracy. We recognize that some language used does not reflect current best practices in reporting demographic characteristics

Themes

NCM support and availability to PCPs

Across the six agencies, NCMs contributed to many aspects of buprenorphine treatment and patient care. These aspects included activities with patients such as providing education about buprenorphine, conducting screenings and intakes, engaging with patients between appointments, providing counseling and emotional support, reviewing urine toxicology results, providing clinical support during titration of initial buprenorphine dose, scheduling appointments with PCPs, identifying pharmacies that consistently stocked buprenorphine, and managing prior authorizations. One PCP highlighted the extent to which the NCM’s role was invaluable:

Because even though I have a big pool of patients who may need the service [buprenorphine treatment], she [NCM] is able to make it much, much, much, much easier. She is able to get the patient, screen them, make sure their insurance, and everything, is in place. And that the patient is a real candidate for the program. And then [she] schedule[s] them for me.

PCP 1, MD, Time 1, Agency 3.

Overall, the support NCMs provided led to tangible benefits for PCPs and their practices. Delegating certain time-consuming aspects of buprenorphine care to the NCM (e.g., intakes) freed up time in PCPs' schedules for appointments, allowing for an increase in their buprenorphine patient panel. For instance, early on, a PCP at Agency 1 noted that sharing the responsibility of care coordination with the NCM had allowed her to see more patients, and thus to apply for the waiver to increase her buprenorphine patient panel from 30 to 100.

In addition, NCM support facilitated enhanced patient engagement in treatment. NCMs and PCPs were able to spend more time with patients during appointments and NCMs were able to efficiently follow up with them as needed. Ultimately, PCPs felt that improvements related to patient engagement and communication were associated with greater rapport and patient-centered care.

[The NCM]'s able to spend more time with them [patients] than we would be in our limited patient slots. So that's been extremely helpful. She's got a great rapport with them, is able to field calls from them in a more timely [manner] than would be done through another system like the front desk. And with a patient group who has so many barriers already, adding even a few more barriers can result in almost invariable [sic] loss to follow-up.

PCP 1, MD, Time 2, Agency 1.

PCPs also highlighted the unique benefits of having a registered nurse coordinate buprenorphine care, given nurses’ clinical training, subject matter expertise, and experience working directly with PCPs. PCPs mentioned multiple times that the clinical support they received from NCMs was superior to that provided by staff members who did not have the same training, expertise, and experience as nurses (e.g., recovery coach, front desk, social workers).

I think that a nurse is more oriented toward patient rapport building and […] making sure that the treatment message from the program gets to the patient in a more digestible fashion and that basic things that the patient is reporting is making its way to the physician. So not only is she […] facilitating communication between the patient and the prescriber, she's also adding in […] her own flavor of clarification and education.

PCP 3, MD, Time 1, Agency 1.

At four agencies, at least one PCP mentioned that the NCM was particularly knowledgeable about substance use disorders and buprenorphine care, including treatment initiation, and two PCPs acknowledged that their agency’s NCM might have been more knowledgeable about buprenorphine care than the PCPs themselves. Furthermore, the positive relationship between NCMs and patients was seen as having a favorable impact on the relationship between PCPs and patients. Multiple PCPs noted that without the clinical support of the NCM, they would not have decided to begin or continue prescribing buprenorphine.

Interviewer: Do you feel like now that you've had that […] time to adjust and have that support with the nurse care manager model that you could continue prescribing on your own without the nurse care manager?

MD: I probably wouldn't. I just can't do this induction on my own, the follow up. Honestly, if the program was not there, I would not take any patients.

PCP 2, MD, Time 2, Agency 2.

While broadly speaking, NCMs were involved in multiple aspects of buprenorphine care, as previously discussed, the type and extent of NCMs’ support to PCPs did differ by agency. One reason for this was variations in initiative implementation. Across agencies, tasks and responsibilities were allocated differently to PCPs versus NCMs. For instance, the NCM at Agency 2 was fully responsible for overseeing buprenorphine treatment initiations (with one PCP at this agency making it clear that they themselves lacked the knowledge and experience to do this), while at Agency 3, PCPs, and not the NCM, oversaw treatment initiations at both time points. Another example of variations in program implementation was the extent to which NCMs participated in efforts to connect buprenorphine patients to additional support services. Some NCMs referred patients to such services while others did not. The reasons for this variability likely depended on multiple factors, including NCM capacity. In several situations, the NCM was assigned to multiple clinics and was unable to provide the level or quality of clinical support that PCPs desired or needed. Many PCPs noted that the NCM’s availability was too limited (both on- and off-site), and some felt that having back-up staff members or a second NCM would be helpful.

Interviewer: And what do you think some of the challenges of the nurse care manager model have been?

NP: Not being here when we need her because she has to go to different sites. So, if we had her five days a week, I think more patients would be here. But, you know, we have other patients we are caring for. So, we don’t get a chance to really sit down if a patient comes and wants to get Suboxone. […] So, so it’s hard. It’s a challenge there. So, I’ve asked her yesterday can she give us more time here?

PCP 3, NP, Time 2, Agency 6

PCPs' comfort with, confidence in, and approach to prescribing buprenorphine

PCPs across agencies reported an improvement over time in their comfort with and confidence in prescribing buprenorphine to people who use drugs through their participation in BNCMI, including PCPs who were initially concerned about buprenorphine prescribing. One PCP described early anxiety around buprenorphine prescribing regulations and fears of surveillance:

Maybe I was a little nervous in the beginning. You know, I thought I was afraid of the rules and regulations and that kind of stuff but, nobody's watching you. You're not going to hurt anybody. I mean you know the guidelines. […] You can learn together […] You know [what] works best.”

PCP 5, PA, Time 2, Agency 1.

Aspects of buprenorphine care that PCPs reported becoming more comfortable with over time included overseeing same-day and home buprenorphine initiations, recognizing opioid withdrawal symptoms, communicating effectively with patients, seeking clinical guidance from colleagues, and interpreting urine toxicology results.

Moreover, some PCPs noted that their approach to care became more adaptable and less judgmental over time, in closer alignment with the principles of harm reduction. For instance, several PCPs noted diminished concern about buprenorphine diversion in parallel with an increased focus on supportive strategies for retaining patients in care.

NP: I'm sure I'm fine tuned. I can't say I'm exactly the same as a year ago. I'm sure –– I think I get better every time. Yeah, I think things get better with time.

Interviewer: What do you think gets better?

NP: Helping someone feel more comfortable or open. I almost never discharge someone, but I'll have more frequent visits with them. Just more confident in what the next step is. I try not to make patients be punished based on my preconceived notions.

PCP 2, NP, Time 2, Agency 3

Relationship between NCMs and PCPs

Overall, NCMs and PCPs reported having positive professional relationships with each other, whether they described their relationship as hierarchical or non-hierarchical. Where tensions existed, they largely stemmed from limited NCM availability (as previously described) and differing approaches to buprenorphine care. For example, during interviews at both time points, the NCM and PCPs at Agency 4 described disagreeing about the appropriate timing of buprenorphine treatment initiation and their response to patients who returned to substance use while in treatment, wherein the NCM favored a more flexible and harm reduction-oriented approach than the PCPs. Similarly, the NCM at Agency 6 noted that one of their four PCPs was not amenable to the harm reduction-oriented, flexible, and patient-centered approach to care espoused by BNCMI.

I feel like we've got a pretty good relationship. I feel like it's pretty collaborative. I think three of the four [PCPs] are really happy to see [buprenorphine] patients. I think one is not particularly interested in it and I've seen that reflected in her numbers. I think she is a little more socially conservative and I think she has different ideas about people’s ability to take responsibility over certain aspects of life. I would guess she is probably somebody who falls a little more into like, “You chose to become addicted, so you can choose to get going out of it.” So, when you're taking that perspective, it's really easy to find reasons to want to discontinue people, feel really frustrated with the patients and feel like it's not working.

NCM, RN, Time 2, Agency 6

Integration of buprenorphine care into primary care

Study participants highlighted multiple benefits of integrating buprenorphine treatment into primary care through OBOT. They noted that primary care clinics are structured to provide care in a way that is less demanding and restrictive than opioid treatment programs (e.g., not expecting patients to come in daily for treatment). Normalizing access to buprenorphine treatment in a setting not traditionally focused on substance use disorders, such as a primary care clinic, allowed for the adoption of a more flexible, holistic, and patient-centered approach to care and excluded the necessity for additional travel to multiple appointments. Lastly, providing buprenorphine in a setting not exclusively devoted to substance use allowed for patients to receive treatment in a more discrete and less stigmatizing manner.

Having [opioid use disorder treatment] in a setting like this is more beneficial because it's, you know, it eliminates the stigma of them having to go to the methadone program. Or some people go to other neighborhoods to get treatment, like, really far away from their house, and it's a major part of their life is their trip. They're just tied down to that. I mean, I think this should be a lot more beneficial and more successful to people who really want help, you know, because - in a primary [care setting] it eliminates all of that.

PCP 5, PA, Time 1, Agency 1

Some agencies offered integrated mental health services to address underlying and co-occurring mental health issues. NCMs and PCPs at these agencies noted the benefits of in-house multidisciplinary support (e.g., access to a psychiatrist, a therapist, or a clinical social worker) for buprenorphine patients experiencing mental health issues.

[O]ne of the advantages that we have here, by being a multi-faceted health community center, is that if I have somebody that has a history of mental health illness, I can call in the doctor right away here, and have an assessment on the spot, and have him sit down.

NCM, RN, Time 2, Agency 4

Mentorship

Participating PCPs and NCMs received mentorship through a variety of sources. These included a formal clinical mentor who was assigned to each agency (an external clinician with expertise prescribing buprenorphine) and a Learning Community which consisted of quarterly meetings where all BNCMI-participating NCMs and PCPs gathered to discuss a topic related to prescribing buprenorphine (e.g., harm reduction, home initiations, racial equity, polysubstance use, or urine toxicology). Learning Community meetings were led by substance use subject matter experts who fostered discussion among staff members across clinics and provided an opportunity for participants to share their experiences caring for patients.

Participants at several agencies mentioned the important role played by the assigned clinical mentor. At Agency 4, the NCM described how the clinical mentor supported them by helping shift a PCP's initial resistance to same-day prescribing. The NCM noted that the mentor's advice carried more weight than the NCM’s advice, given the mentor’s and PCP's commensurate levels of medical training and status as PCPs. When asked how conflicts around differences of opinion between PCPs and NCMs had evolved over the past year, the NCM at Agency 4 noted:

[W]ithout a doubt, you know, yes there’s some apprehension because it [buprenorphine prescribing] was new to them [PCPs]. They don't have experience. Everybody's concerned about, you know, I'm not familiar with this. I don't know how this works. I've never worked with Suboxone...I gotta say that in the past, especially the past four or five months, there's been a better understanding of consulting with each other, and trusting in each other as to -- you know, it's not just because my letters are MD at the end that you have to listen to me because yours are RN only. You know, there's been a little bit of putting each other on the same platform and being open to listening. And I brought in [clinical mentor] a couple of times on conference calls on the spot. And because she has a certain approach, which I love. Because she believes that, you know what? Let's treat ’em all.

NCM, RN, Time 2, Agency 4

Although both NCMs and PCPs appreciated the clinical mentors overall, there was at least one instance of a significant disagreement between a mentor and a clinical staff member. The NCM at Agency 6 expressed her frustration with how the mentor's philosophy did not align with the harm reduction, low-barrier approach of BNCMI.

Our physician mentor says the opposite [of the learning community]. So, it is very frustrating for me. If there is any feedback I might be able to give. To have an extremely conservative physician mentor, who then annihilates that message -- and, so I would like to give feedback that it would be helpful to have someone, if what we’re doing is harm reduction, to have a physician mentor that has a harm reduction approach.

NCM, RN, Time 1, Agency 6.

The above excerpt illustrates that even among experienced buprenorphine prescribers such as the BNCMI clinical mentors, there was still some level of resistance to adopting more flexible, nonjudgmental, and patient-centered approaches to OBOT. Ultimately, it was recognized that this particular physician mentor at Agency 6 was not a good fit for the initiative, and they were replaced by a more harm reduction-oriented mentor before the second interview.

While NCMs generally participated in the Learning Community meetings and described them as helpful, successfully engaging all of the PCPs was challenging. Some PCPs expressed difficulty in finding time in their schedules to attend the sessions and one PCP was still not aware of the existence of the Learning Community at the time of the second interview.

Informal mentorship naturally developed when less experienced PCPs sought advice and support from more experienced PCPs within their own clinic. Agency 3 developed an internal group that was very similar to the Learning Community, wherein PCPs who were prescribing buprenorphine at their institution could share experiences and learn from one another. At some agencies, PCPs described informal mentorship among PCPs as particularly valuable, more so than the assigned clinical mentor or the Learning Community.

PCPs' need for mentorship also evolved over time and differed by agency. At Agency 2, for example, one PCP described an increased need for mentorship as more nuanced clinical questions arose, while at Agency 5, a PCP reported they had reduced their contact with their assigned mentor as they became more comfortable prescribing buprenorphine.

Impact of BNCMI on professional development of participating NCMs and PCPs

Several PCPs noted that a major draw to participating in BNCMI was the opportunity to become an expert in substance use, opioid use disorder, and/or buprenorphine treatment specifically. Further, participating in BNCMI was a critical factor for one PCP to pursue their board certification in Addiction Medicine as a next career step:

As a physician […], [participating in BNCMI] really made me feel, what an exciting time to be a primary care doctor and I’m actually going to sit for the Addiction Medicine Boards. So, part of the qualifying credentials is you have to accumulate enough direct care hours in education and research and administrative hours on topics of addiction and I’ve really been able to do that partially also because my involvement in this grant[-funded] program.

PCP 3, MD, Time 2, Agency 1

In addition, three NCMs noted that providing buprenorphine through BNCMI was a meaningful and rewarding experience that had served as an impetus to pursue additional education to become a Nurse Practitioner in order that they could then prescribe buprenorphine themselves:

That's why I'm striving to study, and continue to study, and hopefully -- become a Nurse Practitioner. Maybe before I'm 60. Not too far off. And that way, I can pick my approach, and, and be able to, you know, have a little bit more liberty, if you will, to take the approach that [clinical mentor] takes.

NCM, RN, Time 2, Agency 4

Continuity and sustainability of BNCMI

Participants reflected on challenges related to the sustainability of BNCMI, in terms of ensuring its continuity during, and viability beyond, the grant funding period. For example, NCMs and PCPs perceived multi-level organizational buy-in (including buy-in from fellow PCPs and from leadership at their agency) and staff member collaboration as two main factors that impacted program sustainability.

The NCM at Agency 2 described how a lack of buy-in from leadership had prevented additional PCPs from becoming waivered at their agency, which ultimately limited the expansion of BNCMI at Agency 2. This NCM shared that while leadership wanted additional buprenorphine patients to increase revenue, they were not providing the necessary backing and paid time for PCPs to attend a buprenorphine waiver training:

RN: The biggest challenge [with BNCMI in the past year] is that the management has no interest in supporting [it]. I just feel that there has not been buy-in whatsoever by the management.

Interviewer: And what do you think their hesitancy is?

RN: I think there's some -- certainly some stigma. I think that people are overwhelmed, and can't focus on what needs to be done. So, they're very poor at making decisions. [...] Like, I think having all of our providers waivered to prescribe buprenorphine -- that is done by leading. And, it's just like not -- was never implemented. There was no leadership with it. I feel like we don't have enough prescribers that [prescribe buprenorphine]. I don't feel that there's stigma at the clinic level, whatsoever. It’s more like the higher-ups. You know, and I still, to this day, have one person [in management] every time we meet say to me, “Well, you know this doesn't work. You know this doesn't work in primary care.” There's that kind of resistance. And, and misunderstanding about like, well, “Why don't you have 200 patients?,” 'cause we've been prescribing. And it's like, it doesn't work that way. They didn't seem interested in pushing [more PCPs to obtain waivers].”

NCM, RN, Time 2, Agency 2

Suggestions for improving buy-in among fellow PCPs within their agencies included: providing PCPs with more clinical mentorship, educating PCPs about stigma towards people who use drugs, encouraging greater participation in the Learning Community, and ensuring that each NCM had the capacity to provide as much clinical support as each PCP would like to receive.

Participants also perceived collaboration with other staff members (e.g., psychiatrists, pharmacists, front desk staff, recovery coaches, non-BNCMI nurses, social workers) as essential to the continuity of BNCMI. Collaboration among clinic staff was particularly important when the NCM was on leave or working at another clinic. However, even with multiple staff providing coverage, several agencies experienced major challenges when the NCM was not available. For example, PCPs were not always familiar with tasks typically carried out by the NCM (e.g., prior authorizations), and other nurses did not have enough time to integrate buprenorphine care into their daily activities.

I’d say the big take-away here is that an average physician cannot cover a nurse’s job when we’re already swamped. But the average nurse can’t cover the [buprenorphine] nurse care manager’s job, because [...] we train these nurses. They’re highly educated in this topic of addiction management more than their colleagues ever will be in the primary care setting, so in covering for a nurse who’s on maternity leave, yes, I have a physician who can help me, and I’m a physician too, but we ideally would’ve wanted a nurse to do it.

PCP 3, MD, Time 2, Agency 1.

In terms of longer-term sustainability, participants noted that without funding from NYC Health Department, their clinics would either no longer be able to afford a nurse solely dedicated to buprenorphine care or would need more patients receiving buprenorphine treatment to justify funding the position. Participants felt that losing the support of the NCM would disincentivize PCPs prescribing buprenorphine. Additionally, participants emphasized that buy-in from leadership was crucial to sustaining the initiative, including through the development of a workable billing structure for NCMs and/or other nurses involved in buprenorphine care.

Discussion

Overall, results from this study demonstrate that both PCPs and NCMs valued BNCMI, a primary care and NCM-based model for buprenorphine care, as an initiative that bolstered their capacity to treat patients with OUD and enhanced opportunities for patient engagement. Participating PCPs and NCMs valued many core aspects of the initiative, notably the comprehensive clinical support NCMs provided, and the benefits of integrating buprenorphine treatment into a primary care practice.

Study participants also considered the cross-agency Learning Community meetings and the guidance provided by the assigned clinical mentor at each agency as important elements of the program. However, challenges were noted at multiple agencies related to NCMs’ limited capacity, conflicting approaches to treatment, lack of buy-in at various levels, and concerns around sustainability.

The Diffusion of Innovation Theory provides a useful framework for interpreting these data [21]. Mapping study results against key concepts from Rogers’ theory allows for a discussion of BNCMI elements in light of factors known to enable or hinder diffusion of an innovation [26–28]. The five main factors posited to influence adoption of an innovation are “relative advantage” or the extent to which an innovation is seen as better than the program it replaces, “compatibility” or the extent to which an innovation is consistent with the values, experiences, and needs of the potential adopters, “complexity” or the extent to which an innovation is easy to understand or use, “trialability” or the extent to which an innovation can be tested prior to adoption, and “observability” or the extent to which an innovation’s effects can be observed. The theory also emphasizes the importance of “change agents” and “opinion leaders” as key players in supporting and leading others in the adoption of new models, and it classifies “consequences” of an innovation as desirable versus undesirable, anticipated versus unanticipated, and direct versus indirect.

Findings from this study demonstrate that generally, participants described BNCMI as having very high relative advantage, mostly high compatibility, mostly low complexity, high observability, and mostly low trialability.

Staff largely saw very high relative advantage to participating in the initiative. PCPs and NCMs viewed the integration of OBOT into primary care as a way to reduce stigma and expand access to life-saving treatment, highlighting the value of harm reduction principles embodied in OBOT itself. Many participants reported that the structured clinical mentorship and quarterly Learning Community meetings offered by BNCMI were helpful. Staff also overwhelmingly felt that registered nurses were the most appropriate members of the patient care team to take on the duties and responsibilities encompassed in the role of NCM. Critically, multiple PCPs noted that they would not have begun or continued to prescribe buprenorphine without the clinical support provided by the NCM.

Most NCMs and PCPs described the approach and components of BNCMI as mostly highly compatible with their values, needs, and experiences as providers. Study participants believed in expanding access to buprenorphine and saw safety-net primary care clinics as a suitable setting to reach and treat people with OUD with buprenorphine. As PCPs gained more experience and became more comfortable prescribing buprenorphine, many described their practices as evolving towards a more harm reduction-oriented, flexible approach to buprenorphine care, as encouraged by the NCMs and at the Learning Community meetings.

However, some participants described instances of incompatibility between BNCMI and individual PCPs' approaches to care. For instance, some PCPs did not embrace the principles of harm reduction that were supported by the BNCMI initiative and disagreed with NCMs about certain treatment-related decisions. Whereas NCMs generally emphasized the need to be flexible and supportive with patients who continued or returned to using drugs during treatment, some PCPs, and at least one mentor, favored more punitive strategies. PCPs varied in their approaches to implementing low-threshold OBOT; while some were fully onboard with the principles of harm reduction, others resisted adopting certain aspects of providing flexible, nonjudgmental, and patient-centered care.

Some participants also described incompatibility with leadership priorities. At some sites, leadership did not always provide adequate time for their PCPs to receive the necessary training to obtain a buprenorphine prescribing waiver, hindering the expansion of BNCMI services within their organization. There were also mentions of incompatibility of BNCMI with organizational practices, protocols, and staff workloads; for example, several PCPs lacked time in their schedules to attend Learning Community meetings.

Although PCPs initially perceived first-time buprenorphine prescribing for primary care providers as highly complex (e.g., learning to appropriately titrate buprenorphine during treatment initiation and interpreting urine toxicology results), PCPs' perceptions of BCNMI complexity decreased over time as they gained more hands-on experience and received continuous mentorship and NCM support. As such, ultimately PCPs saw the initiative as having mostly low complexity.

BNCMI can be considered as having high observability, as there were many opportunities for NCMs and PCPs to directly see the extent to which BNCMI brought tangible results. Study participants described how participating in BNCMI noticeably benefitted their own practice, for example by expanding their expertise and interest in continuing to treat substance use disorders. They also noted seeing first-hand how integrating buprenorphine into primary care allowed for more PWUD to access substance use treatment in a non-stigmatized setting and how collaborative coordination of buprenorphine treatment among NCMs and PCPs enhanced patient engagement. As discussed in a complementary qualitative study published in 2022, patients who were receiving or had received buprenorphine treatment through BNCMI also reported observing many positive outcomes from the initiative. They noted that integrating buprenorphine treatment into primary care destigmatized substance use treatment, improved retention in treatment, and provided access to other medical care that could improve overall health 30.

BNCMI offered mostly low trialability since there was no opportunity to test out the program prior to commitment to participation. At each agency, organization leadership applied to receive funding to participate in BNCMI through a formal grant application prior to the start of the initiative. For individual PCPs, trialability was also limited because they were required to obtain a prescribing waiver (which required training) prior to participating in BNCMI at their agency, thus necessitating at least minimal individual commitment to the initiative prior to testing it.

Two key aspects of study participants’ experiences with BNCMI were the dynamics among PCPs, NCMs, and mentor(s), and the individual roles they each played. NCMs were, by definition, envisioned as the BNCMI change agents. NCMs sat midway between their agency and NYC Health Department (employee of their agency but funded by NYC Health Department, expected to attend all NYC Health Department-led technical assistance and mentorship activities and guide new PCPs), regularly interacted with BNCMI clinical mentors, and overall were central to the implementation of buprenorphine treatment delivery at their respective agency. At many agencies, NCMs fulfilled criteria described by Rogers as factors for successful change agents: being empathetic, making efforts to communicate regularly with PCPs, understanding PCPs' needs, being perceived as trustworthy by PCPs, and being in close contact with BNCMI mentors. However, as discussed in the results, these factors were not always present, such as when NCMs lacked capacity to provide the consistent clinical support PCPs needed, or when PCPs did not feel confident enough in the NCMs’ clinical background as RNs to consider the NCMs’ opinions during disagreements.

PCPs relied heavily on other clinical peers (the assigned clinical mentor, other buprenorphine PCPs at their agency or from the Learning Community) for clinical guidance and mentorship. The clinical mentors and peers (in Diffusion of Innovation terms, opinion leaders), were trusted for their insight and experience and contributed to BNCMI success. Overall, it appears that PCPs felt most comfortable, and the initiative progressed most, when both change agents and respected opinion leaders were present, provided consistent messages, and worked together.

Concerns around the sustainability of BNCMI beyond the NYC Health Department-funded grant period, namely its financial sustainability, emerged as a major theme across agencies. According to Rogers' theory, the sustainability of an innovation relies on several factors: stakeholder involvement during its implementation, the extent to which the organization modifies and adapts the innovation as it diffuses, the presence of a local champion, and the alignment between the innovation and the organization. Though NYC Health Department continues to fund and provide technical assistance for BNCMI in NYC, care coordination for substance use provided by RNs is still not a billable service in New York State, and therefore agencies are reliant on grant funding to provide this model of care. In addition to funding, NYC Health Department has been supporting BNCMI sustainability by providing ongoing education and training when there is staff turnover and developing contractual requirements for agencies to devote resources to new patient engagement.

Rogers classified the consequences of innovations as desirable versus undesirable, direct versus indirect, and anticipated versus unanticipated [21], providing a useful lens for examining the overall success of BNCMI’s success across agencies. Anticipated desirable consequences included clinicians becoming more comfortable prescribing over time and learning more about buprenorphine treatment and care through the initiative (direct), and organizations being able to better treat (by prescribing buprenorphine), engage, and retain PWUD in care (indirect). There were even unanticipated desirable consequences to the diffusion of BNCMI, as multiple PCPs and NCMs pursued further expertise in substance use treatment.

In a study comparing the process of adopting buprenorphine treatment for OUD across two health insurance plans that previously did not offer buprenorphine treatment, Green et al. also used the Diffusion of Innovation theory to examine the elements that led to successful adoption [28]. While Green et al.’s model and study differs from BNCMI (e.g., BNCMI study participants had all already adopted buprenorphine treatment, whereas in Green et al.’s study only some had adopted it), the authors’ findings support some of the findings from this study. Successful adoption of buprenorphine depended on factors such as advocacy of clinical leadership and PCPs, clinical mentorship and guidance from colleagues with experience prescribing buprenorphine, and clinicians’ own individual experiences observing the positive effects of buprenorphine treatment on their patients. Clinicians had the opportunity to learn from leaders, other colleagues, and their own experiences, which overall increased their comfort prescribing buprenorphine. Other important characteristics described by Green et al. that held true for BNCMI included clinicians’ treatment orientation and philosophy of treatment; notably, clinicians who endorsed harm reduction approaches to care were more likely to support use of buprenorphine treatment.

Using the Diffusion of Innovation Theory as a framework helps to highlight many of the ways in which BNCMI proved to be successful as an adoptable innovation, as well as areas with room for improvement. Table 2 lays out our recommendations, grouped by theme, for addressing the challenges of BNCMI discussed above. More generally, these are strategies that can be useful to support and enhance the diffusion of buprenorphine prescribing in primary care settings.

Table 2.

Recommendations for addressing BNCMI challenges

Theme	Recommendation(s)
NCM availability to PCPs	Ensure that NCM capacity meets the needs of PCPs at each agency; this might mean that more than one NCM is needed per agency
PCPs' comfort with, confidence in, and approach to prescribing buprenorphine	Provide comprehensive clinical support and guidance to new buprenorphine PCPs for at least 12 months following commencement of program implementation. This will allow PCPs to build confidence in treating people with OUD and increase their comfort level prescribing buprenorphine through hands-on experience. •PCPs should receive clinical support from the NCM, formal mentorship and guidance through the Learning Community and the assigned mentor, in addition to informal clinical mentorship and guidance from other experienced PCPs. Implement activities early on to build consensus and agreement around a shared understanding of harm reduction principles among all BNCMI providers (NCMs, PCPs, and mentors). •Future studies should examine providers’ attitudes around PWUD and harm reduction principles at baseline and follow-up. All participating PCPs and NCMs should utilize a mutually agreed-upon framework for providing buprenorphine treatment, based on current pharmacologic and harm reduction literature, and have shared practices around dosing decisions and patient engagement.
PCP-NCM relationship	Recognize that disagreements might occur between PCP(s) and NCMs, especially in the beginning of the implementation of BNCMI. Build structures and protocols for navigating disagreements, such as having NCMs and PCPs consult with mentors or other PCPs who prescribe buprenorphine. PCPs and NCMs should consult assigned clinical mentors regularly and participate in the Learning Community to discuss and receive guidance around issues that arise. Emerging topics of disagreement can be discussed in Learning Community meetings for broader consideration. To enhance care provision, implement interdisciplinary team-based models with clear role definitions. For long-term success, focus on fostering respect and collaboration across the team, rather than relying on individuals whom PCPs already respect to resolve disagreements.
Clinical mentorship	Offer high-quality formal and informal mentorship to PCPs (e.g., assigned mentors, Learning Community, relevant trainings). Ensure that approach to care is consistent across assigned mentors and Learning Community for clear and consistent messaging to BNCMI PCPs. Facilitate mentorship within individual clinics among new and more experienced PCPs.
Impact of BNCMI on professional development of participating NCMs and PCPs	Encourage and support NCMs to pursue advanced nursing degree (Nurse Practitioner) and become PCPs themselves; BNCMI can be viewed as a pipeline for professional development.
Continuity and sustainability of BNCMI	Ensure that all BNCMI-involved staff are aware of BNCMI protocols at the start of implementation and iteratively develop and document protocols to share with new NCMs and PCPs. Establish a contingency plan for when NCM is on leave or the role is temporarily vacant, to ensure continuity of patient care. •Train other clinic staff, such as Licensed Practical Nurses (LPNs) and Registered Nurses (RNs), to provide coverage for the NCM when needed. Establish sustainable funding structures and upper-level management/leadership buy-in for long-term program sustainability and fiscal viability.

Limitations

Because data for this study were only collected at two times during the first year of BNCMI at each clinic, we do not know how BNCMI providers’ perspectives, practices, and dynamics evolved after the first year. Future studies should explore later stages of implementation and sustainability of similar initiatives for further insights. Additionally, it is possible that our findings and recommendations may not generalize to contexts other than urban, primary care safety-net health care systems with similar levels of resources and infrastructure. This study is also limited in that it only includes provider perspectives and does not incorporate the voices or experiences of patients. Additionally, we do not have quantitative data to assess the direct impact of the intervention on patient health outcomes. Finally, because BNCMI was a relatively small-scale initiative, it is not possible to attribute population-level changes in morbidity or mortality rates to the program. Together, these limitations highlight the need for caution when applying the results more broadly, and underscore the importance of continued evaluation in shifting contexts.

On the other hand, trustworthiness of the findings was enhanced by the fact that none of the researchers (including interviewers) were involved in BNCMI implementation, which allowed participants to demonstrate openness and comfort in sharing their challenges or concerns with the initiative.

Conclusion

This qualitative process evaluation of the NYC BNCMI showed that the concomitant provision of NCM clinical support, clinical mentorship, and learning opportunities for PCPs new to prescribing buprenorphine increased PCPs' willingness and capacity to offer buprenorphine treatment in safety-net primary care settings. Essential elements of a successful primary care-based buprenorphine NCM initiative include sufficient availability of and consistent support from an NCM to PCPs; robust, varied, and consistent mentorship structures; leadership buy-in; and funding plans for ongoing fiscal viability. Initiatives such as BNCMI can improve PCPs' ability to provide high-quality patient-centered care to people with OUD, expand access to life-saving treatment, and ultimately contribute to stemming the opioid overdose crisis.

Abbreviations

BNCMI

Buprenorphine nurse care manager initiative

LPN

Licensed practical nurse

MD

Medical doctor

NCM

Nurse care manager

NP

Nurse practitioner

NYC

New York City

OUD

Opioid use disorder

PA

Physician assistant

PCP

Primary care provider

PWUD

People who use drugs

RN

Registered nurse

Author contributions

NB participated in the data curation/acquisition for this study. NB, ECW, MKD, and AH conceptualized the study. NB, ECW, and AH developed the methodology for the study. NB and ECW analyzed the data and co-led on the investigation. AH supervised the investigation. ECW oversaw and managed the administrative aspects of the manuscript development process. NB and ECW developed the original draft of the manuscript. AH, MKD, and ATR reviewed the manuscript, and ECW led on making edits. All authors read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data used in this study contain sensitive information and cannot be shared publicly. However, de-identified data may be available upon reasonable request from the authors.

Declarations

Ethics approval and consent to participate

This study was approved by the NYC Health Department’s Institutional Review Board. All participants provided written informed consent before participating in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.