Improving access to buprenorphine for rural veterans in a learning health care system

Abstract

Objective

To describe a learning health care system research process designed to increase buprenorphine prescribing for the treatment of opioid use disorder (OUD) in rural primary care settings within U.S. Department of Veterans Affairs (VA) treatment facilities.

Data Sources and Study Setting

Using national administrative data from the VA Corporate Data Warehouse, we identified six rural VA health care systems that had improved their rate of buprenorphine prescribing within primary care from 2015 to 2020 (positive deviants). We conducted qualitative interviews with leaders, clinicians, and staff involved in buprenorphine prescribing within primary care from these sites to inform the design of an implementation strategy.

Study Design

Qualitative interviews to inform implementation strategy development.

Data Collection/Extraction Methods

Interviews were audio‐recorded, transcribed verbatim, and coded by a primary coder and secondary reviewer. Analysis utilized a mixed inductive/deductive approach. To develop an implementation strategy, we matched clinical needs identified within interviews with resources and strategies participants had utilized to address these needs in their own sites.

Principal Findings

Interview participants (n = 30) identified key clinical needs and strategies for implementing buprenorphine in rural, primary care settings. Common suggestions included the need for clinical mentorship or a consult service, buprenorphine training, and educational resources. Building upon interview findings and in partnership with a clinical team, we developed an implementation strategy composed of an engaging case‐based training, an audit and feedback process, and educational resources (e.g., Buprenorphine Frequently Asked Questions, Rural Care Model Infographic).

Conclusions

We describe a learning health care system research process that leveraged national administrative data, health care provider interviews, and clinical partnership to develop an implementation strategy to encourage buprenorphine prescribing in rural primary care settings.

What is known on this topic

• Addressing opioid use disorder (OUD) is an urgent clinical priority in the United States, yet a minority of patients receive gold standard treatment, medications for opioid use disorder (MOUD).

• While access is suboptimal for all patients, rural patients are particularly underserved.

• Buprenorphine, a form of MOUD that can be prescribed in primary care, has potential to expand access for rural patients, but primary care clinicians often remain reticent to prescribe.

What this study adds

• In this study, we sought to develop a strategy to overcome barriers to the integration of buprenorphine within rural, primary care settings by leveraging VA's capacity as a learning health care system.

• Interview participants within rural health care systems that had improved their use of buprenorphine identified the need for clinical mentorship and consultation, training, and educational resources to support buprenorphine prescribing in primary care.

• Building on these findings, we developed an implementation strategy composed of connection to a consult service, a process of audit and feedback, training, and educational resources.

1. INTRODUCTION

Addressing opioid use disorder (OUD) remains an urgent clinical priority in the United States. An estimated 2.0%–2.8% of the US adult population has OUD, and the presence of fentanyl, and more recently xylazine, in the drug supply have increased the harms associated with opioid use. ¹ , ² , ³ In 2021, 81,000 Americans died from an opioid‐related overdose. ⁴ Despite the serious implications of OUD, only a minority of those with the disorder receive medication treatment for OUD (MOUD). ⁵ , ⁶ MOUD, which includes formulations of methadone, buprenorphine, and naltrexone, is considered first‐line, “gold standard” treatment. ⁷ , ⁸ , ⁹

While patient access to MOUD has improved, ¹⁰ , ¹¹ MOUD remains less accessible for rural patients, for whom geographic distance to specialty substance use disorder (SUD) treatment programs, opioid treatment programs, and MOUD prescribers may be prohibitive. ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ To expand rural patients' access to MOUD, health systems have frequently focused efforts on buprenorphine, a medication that can be prescribed within primary care, in contrast to methadone, and that is more readily acceptable to patients, in contrast to naltrexone. ¹² , ¹⁷ , ¹⁸ , ¹⁹

Prior research has identified barriers and facilitators to expanding access to buprenorphine in rural and primary care settings. Barriers identified have included perceived lack of clinician time and resources, lack of mental health or psychosocial support services, lack of clinician training and confidence, financial disincentives, stigma, and concerns about diversion. ¹⁶ , ²⁰ , ²¹ , ²² , ²³ Conversely, identified facilitators include the use of a team‐based or integrated care approach, leadership support, tele‐prescribing, financial resources and incentives, and the use of discrete implementation strategies (e.g., external facilitation, tele‐education, learning collaboratives, training, and technical assistance). ²⁰ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰

Within the U.S. Department of Veterans Affairs (VA), the largest integrated health care system in the United States, expanding access to MOUD for all patients, and rural patients specifically, is a major clinical priority. ¹² , ¹⁸ , ³¹ , ³² Although the gap in MOUD accessibility between rural and urban VA patients has declined over time, ¹⁴ , ³³ rural patients continue to face barriers to accessing treatment. ³⁴ One key barrier is the centralization of addiction treatment services within specialty SUD and mental health clinical settings, as opposed to primary care, which is more readily available in the community‐based outpatient clinics where rural VA patients frequently receive care. ¹² , ³⁵

To overcome barriers to the integration of buprenorphine within rural, primary care settings, we leveraged VA's capacity as a learning health care system. ³⁶ , ³⁷ , ³⁸ A learning health care system is characterized by a culture of learning and improvement, reliance on data and evidence to improve care, and partnerships between researchers, clinicians, leaders, and patients. ³⁹ , ⁴⁰ In a learning health care system, research questions are informed by clinical and leadership priorities as well as the state of the science, and study results are disseminated back to key stakeholders with the goal of shaping clinical care and policy decision‐making. ³⁷ , ⁴¹

In this study, we describe the process through which, with input from clinicians, leaders, and staff, we developed an implementation strategy to increase buprenorphine prescribing in rural primary care settings.

2. METHODS

2.1. Data sources and study setting

This study utilized national VA administrative data from the Corporate Data Warehouse (CDW), which contains patient‐ and facility‐level data for VA, to identify rural VA health care systems that had improved their rate of buprenorphine prescribing within primary care from 2015 to 2020 to sample for qualitative interviews. The clinical setting in which buprenorphine was prescribed was identified by the clinical context of the buprenorphine prescriber. For each year, we calculated the rate of primary‐care‐based buprenorphine prescribing for each VA Health Care System by dividing the number of patients diagnosed with OUD in each system by the number who received buprenorphine in a primary care context within 12 months of OUD diagnosis. Improvement was defined as an increase of approximately five or more percentage points in the proportion of patients prescribed buprenorphine in primary care in one or more study years, and maintained through the final study year. Health systems, which are composed of one or more medical centers and their affiliated community‐based outpatient clinics, were identified as rural based on the Rural–Urban Commuting Area (RUCA) code of the primary VA medical center.

Of 15 rural VA health care systems, we identified six with sustained improvement through 2020; all six were selected for interviews. These systems were located across the United States in the Northwest, West, Midwest (2), Northeast, and South. In 2020, the average number of patients served by each Health Care System was 30,000 (range: 20,000–38,000). In these six systems, on average, no patients received buprenorphine in primary care in 2015 (range 0–0.5%). At that time, patients could receive MOUD in a specialty SUD or mental health setting, or symptoms would go untreated. By 2020, an average of 11.6% of patients diagnosed with OUD received buprenorphine within primary care (range 4.5%–37.1%). In comparison, of the nine rural sites that did not improve rates of primary care buprenorphine prescribing, an average of 0% of patients received buprenorphine in primary care in 2020 (range 0–0.3%).

2.2. Qualitative data collection and analysis

We sought approval to conduct the study from medical center leadership via email. Initial outreach emails informed leaders of their site's status as an “improved” facility, emphasized that expanding access to buprenorphine is a major VA priority, and encouraged participation by highlighting the role of national SUD leadership as operational partners in the research; these partners were carbon copied on outreach emails. All medical center leaders contacted agreed to participate and provided contact information for primary care leaders. Similarly, all primary care leaders contacted agreed to participate and provided the names of clinicians and staff involved in, or knowledgeable about, buprenorphine prescribing in primary care.

From each site, we recruited five interview participants representing distinct roles (e.g., clinical pharmacist, physician, primary care leadership) via email and instant message outreach to participate in a single interview. Email outreach described how sites were identified and the goals of the study. Of 36 potential participants contacted, 30 agreed to participate.

Qualitative interviews and analysis were conducted by two PhD‐level, female‐identifying social scientists (JW and SN), each with 15+ years of qualitative research experience, and a master's level, non‐binary research assistant (AE), trained in qualitative interviewing and analysis. All were VA employees. One participant was a former colleague of an interviewer; all other participants had no preexisting relationship with interviewers. Interviews spanned 30–60 minutes, were conducted by phone or video, and were recorded and transcribed verbatim. Interviewer role (principal investigator, research assistant, analyst, VA employee) was shared within the interview. Semi‐structured interviews informed by Normalization Process Theory probed the processes through which primary‐care‐based buprenorphine prescribing had been established, as well as barriers and facilitators to expansion of buprenorphine in primary care. ²⁹

Normalization Process Theory suggests that successful implementation of a new clinical practice must attend to social and organizational aspects of care delivery, and work to embed or “normalize” the new practice within routine clinical care. This approach is consistent with prior research showing persistently low availability of buprenorphine within primary care despite leadership and staff belief in its efficacy and importance, ⁴² , ⁴³ suggesting that barriers to uptake may reflect challenges integrating the practice within existing workflows (Figure 1).

FIGURE 1.

Conceptual model based on normalization process theory.

Analysis utilized a mixed inductive/deductive approach. Three members of the research team (JW, SN, and AE) independently coded a subset of three interview transcripts using the software program ATLAS.ti. Codes were both deductive, reflecting theoretical constructs, and inductive, reflecting emergent concepts. The team met to discuss differences and come to consensus on a final codebook. All transcripts were coded by a primary coder and reviewed by a secondary reviewer, with differences resolved through consensus methods. To analyze the interview data, we pulled all quotes tagged with codes addressing sites' processes for implementing buprenorphine and advice for other systems. We grouped quotes identified by these codes into categories representing commonly voiced clinical needs and supports for buprenorphine prescribing in primary care. Illustrative quotes are presented in the text and tables. All quotes are identified by the participant's clinical role, a unique numeric identifier, and participants' gender (M/F).

2.3. Implementation strategy development

Building upon these findings, we developed an implementation strategy to be piloted in rural community‐based outpatient clinics within one VA Health Care System. To develop the strategy, we matched key clinical needs identified within interviews with resources and strategies participants had utilized to address these needs in their own sites. To address clinical needs for which our data did not suggest a resource or strategy, we searched for existing electronic or print resources and iterated new resources with input from clinical partners.

Concurrently, two members of the research team, clinical experts in pain, MOUD, and opioid safety (KM, KK), had independently begun to develop a buprenorphine training targeting primary care clinicians in community‐based outpatient clinics within our local VA Health Care System. This training was created in response to requests from primary care clinicians and clinic practice managers for practical information on using buprenorphine to treat OUD and chronic pain. Training was scheduled to be delivered in four community‐based outpatient clinics.

Of these four clinics, we selected two that served a substantial number of rural veterans. Following the trainings, we met with practice managers in these sites and asked for their participation in an implementation intervention and evaluation process (described in detail below). Both clinic managers agreed to participate in the process, which is ongoing.

This study was approved by the Institutional Review Board at the VA Portland Health Care System. Study results are reported consistent with the Consolidated Criteria for Reporting Qualitative Studies (COREQ) guidelines. ⁴⁴

3. RESULTS

To inform the development of the strategy, we completed 30 interviews with leaders, clinicians, and staff across six sites. See Table 1 for sample characteristics.

TABLE 1.

Interview participant characteristics (n = 30).

Gender, n (%)
Female	14 (47%)
Professional role, n (%)
Leadership	5 (17%)
Pharmacist	10 (33%)
Physician	6 (20%)
Registered Nurse	2 (6%)
Nurse Practitioner	5 (17%)
Social worker	2 (6%)

3.1. Clinical needs and supports to facilitate buprenorphine prescribing in primary care

We first describe the clinical needs and supports participants identified as central to integrating buprenorphine prescribing into primary care. (See Table 2.) We then detail components of the implementation strategy we developed to address these needs.

TABLE 2.

Clinical needs and supports for primary‐care‐based buprenorphine prescribing identified by interview participants.

Needs and supports	Illustrative quotes (Role, Study ID, gender)
Consult service/clinical mentorship
Experienced provider database	“The main thing is talk to somebody that has experience in it… Have a database of ‘Hey, if you have questions, reach out to this therapist… If there was somebody like, even a specialist in each [VA region], that would be great.’” (pharmacist, 15, M)
Clinical mentorship/consult Service	“I think setting up where they have access to a mentor, someone who is…senior, experienced in prescribing. Because I have questions all the time… if you're in a [community‐based outpatient clinic] and you're remote and rural and there's nobody else but you…you may not feel so comfortable, secure enough …I think if there was something that was set up that they knew ahead of time, hey, you have good resources that are available to you when you need it to guide you through…review your chart, give you feedback… that could be helpful.” (nurse practitioner, 30, M) “… it'd be difficult if we didn't have someone to mentor or talk through cases … that has certainly helped us here expand that slowly but steadily. Instead of… people doing it and being like, well, now what do I do? I don't have someone to talk through. I don't know where to start.” (primary care leadership, 25, M) “Having mentorship, I mean, I honestly think if there's gonna be a single thing that we should push…nationwide, if we're really serious about making this a widespread treatment available to everybody then there has to be some sort of mentorship available. There needs to be like the equivalent of a consult… either have a consult or have a local or [VA region] based provider who is willing to do, to be that person.” (physician, 26, M)
Training and educational materials
Engaging in‐person training	“The one thing I wish I had a little more [of is] one‐on‐one training and that's gonna be key because honestly, I'm not 100% comfortable myself…” “I really think that education on the provider and staff side is just…a huge piece… either you have to have someone who's really comfortable already and sort of be the model of what it looks like to help engage… people or have an education program in a way that engages people and their comfort level….” (social worker and pain coordinator, 13, F) “It's the human connection, [when] it's on video– people they do their notes, they're signing things, they're not even looking at it, they may or may not register it.” (nurse practitioner, 3, F)
Peer to peer	“Getting someone that's a peer‐to‐peer colleague to come in…is gonna be the most helpful way to persuade folks that are just really stuck… Those sorts of trainings and experiences I think would really be more practice changing.” (pharmacist, 5, F)
Medication guidance (e.g., dosing, initiation)	“When they do the training through the VA, it was good giving you the foundation and the theory and the practicality of it on the basics, but I did…feel like [it] could have been a bit better in the actual prescribing, dosing…equivalents, if you have to switch from one formulation to another because you're going from milligrams to micrograms… even the psych pharmacist felt like it was kind of guessing, in a way…” (nurse practitioner, 30, M) “I think it would be…how to initiate it based on what are they using? This is how many hours you typically wait. Give them this dose. Wait this long, kind of walk them just through the process and then the characteristics of the medication… Just…have like a formal‐ish process that they can follow…” (social worker/pain coordinator, 13, F)
Overcoming medication stigma	“I think most people have hesitancy. There's a mindset that's out there that you're trading one [addiction] for the other – you're taking someone who is basically a heroin addict and they're on pure opium and you're moving them to suboxone, but what is the goal? Are they going to stay on that forever?” (physician, 2, F) “I would say a lot of provider education would be important. I think it took quite a while for…the primary care providers and other providers within the facility to realize that there is a medication that can help these patients and they get the stigma away from it in a way.” (pharmacist, 21, F)
Overlap of pain and opioid use disorder	“So it's not like you can make this diagnosis quickly. Because in most cases it's the patients themselves that don't want to accept it, because they don't want to go off of their pain medicine. It's really incredibly hard… Most of the time I can tell it's OUD, but nobody's going to call it that, they're all going to call it pain.” (primary care leadership, 7, M)
Information on what has worked in other clinics	“Make sure that you look at other clinics… so that you can get a good understanding of what the process is, and then pick and choose from those other clinics to do…what you think is most appropriate.” (nurse practitioner, 22, F) “…trying to figure out how other people have done [it], what resources they used…” (pharmacist, 9, F)

Abbreviations: F, female; M, male; OUD, opioid use disorder; VA, Department of Veterans Affairs.

3.1.1. Clinical mentorship

The most common advice participants gave, and the most common facilitator to participants' buprenorphine prescribing, was connection with an experienced prescriber/clinical mentor to provide guidance, answer questions, and build new prescribers' confidence. Mentorship was needed because many clinicians felt inadequately prepared to begin prescribing buprenorphine, “[primary care providers] feel like they don't have the education. They feel a little bit unsure about the medication… from what I'm understanding, they struggle,” as one clinical pharmacist (9, F) described.

This lack of confidence clashed with clinicians' recognition of the need for OUD treatment among their patients, as well as pressure from local and national leadership. As one participant described, “there's no reason in the world that a rural veteran who walks into a [community based outpatient clinic] and says, ‘I'm withdrawing now, I'm dopesick but I don't want to use, can you help me?’ [should] have to go to the main facility to get treated.” (Leader, 19, M). Distance from community‐based clinics to the nearest medical center could be prohibitive; in our sample, it ranged from 22 to 495 miles. While many clinicians recognized the need for OUD care for their rural patients, clinical mentorship provided the confidence to move forward.

Clinical mentors provided encouragement and support to new prescribers, who were often hesitant to take on a new clinical responsibility. One physician (27, M), who had served as a clinical mentor to new prescribers working in community‐based clinics across the state, noted: “It's clear to me that a little bit of hand holding on my part made a huge difference for them [clinicians in rural settings] in terms of their confidence.”

Mentorship was viewed as particularly important for smaller, outpatient settings where providers could feel isolated in their work. A physician (26, M) described how mentorship could help clinicians in community‐based outpatient clinics successfully transition from interest in prescribing buprenorphine to active participation:

You can do the training…they do a couple of examples and you're like, ‘OK, I sort of see’… It's just different when you're working day‐to‐day and you're burned out and exhausted…It would really be helpful to…have somebody…you could…talk to…review the case…make sure you're not missing something…people would feel more confident about doing the training and actually prescribing if they have that as…a support.

In small practices where there could be just a handful of staff in the building, the ability to consult with an experienced prescriber was highly valued.

Across research sites, clinical mentorship was often provided informally, through conversations or instant messages between clinicians known to one another. However, in one system, the mentoring process was more structured, with the clinical pharmacist acting as a liaison between clinical mentors and new prescribers. New prescribers could contact the pharmacist to be connected with an experienced prescriber (either a primary care clinician or a psychiatrist) to staff the patient, conduct joint visits, or transfer the patient to the experienced prescriber, with primary care remaining involved. The goal of this mentorship was, “simply so that we can ensure that [the new prescriber] is as comfortable as possible.” (pharmacist, 29, M). In this way, clinical mentorship laid the groundwork for establishing buprenorphine prescribing in primary care.

3.1.2. Clinical consultation service

The second most frequently requested resource, addressing the same concerns albeit through a different mechanism, was a clinical consult service. One clinician described how such a service could provide the scaffolding from which a clinician could move from intentions to action:

“…if you think this person has opioid use disorder, [but] I'm very new at this. I want to do the right thing…I really need help…have a consult or…a local or [regional] provider who is willing…to review the chart, help you with management…” (physician, 26, M)

Pairing new with experienced prescribers to review cases and provide feedback on care decisions was viewed as particularly important in the rural context:

“… if you're in a [VA community‐based outpatient clinic] and you're remote and rural and there's nobody else but you…you may not feel…secure enough to say…I think this would be a good option…[if you] knew ahead of time…you have good resources that are available to you when you need it to guide you through…review your chart, give you feedback…that could be helpful.” (nurse practitioner, 30, M)

For rural clinicians who lacked an on‐site clinical mentor to support them through the process of beginning to prescribe buprenorphine, a consult service would help fill the gap.

3.1.3. Education and resources

Interview participants also described the need for education and training around buprenorphine prescribing, including the logistics of treatment and care processes, overcoming stigma, and how to address patients experiencing diagnoses at the intersection of OUD and chronic pain. Some participants felt that an engaging, in‐person didactic training would best meet this need, while others recommended that print and virtual access to reference materials would be more valuable. One clinical pharmacist (29, M) explained, “certainly smaller sites and within my own [VA region] there's just a lot of confusion as to how to go about using buprenorphine, and incorporating buprenorphine and…even our own policies…” Another provider who served in the role of pain coordinator (social worker, 13, F) noted, “I think that particularly in the rural areas you just have less access to newer stuff and so newer stuff comes up, but you don't necessarily take the time to catch up with it until you really have to or it's right in front of you.” While acknowledging that requiring participation in trainings was “not necessarily the best,” this participant felt that trainings paired with education resources could yield greater clinician buy‐in, “you get a lot further versus saying ‘You need to do this. Good luck.’”

Participants described a number of resources and training materials they felt could support primary care clinicians new to prescribing buprenorphine. Suggestions included clarification regarding buprenorphine rules and policy (e.g., educational requirements, patient limits), information on home induction, and standard of practice guidance. Participants thought these resources would help to clarify policy and procedures. They also suggested that education could address false perceptions or stigma, such as the belief that, “oh my gosh, this is a dangerous medication and I don't want to prescribe,” as one clinical pharmacist (14, F) described. Some participants recognized that VA already hosts resources of this sort, but that providers may not always know how to access them: “Having a definable sort of resource…and place for people to turn would be good for us as a system…[VA is] tremendous about offering 1000 different resources. But they can also be hard to find.” (clinical pharmacist, 29, M).

3.2. Implementation strategy components

Drawing upon advice from interview participants, we designed an implementation strategy to support buprenorphine prescribing in rural, primary care settings. The components of the strategy included connection to existing consult services, a process for delivering audit and feedback, a buprenorphine training led by clinical partners, and educational materials. Table 3 pairs clinical needs, summarized with illustrative quotes, with components of the strategy targeting these needs.

TABLE 3.

Resources frequently requested and utilized by interview participants and targeted implementation strategy components.

Requested resources	Illustrative quotes (Role, study ID, gender)	Implementation strategy component
Clinical mentorship/Consultation service	“I think setting up where they have access to a mentor, someone who is… senior, experienced in prescribing. Because I have questions all the time.” (nurse practitioner, 30, M) “If there was anything that would help it would be [an] opioid use disorder consult.” (physician, 26, M) “…there's definitely a hands‐on training and discussion component. But there is [also] a component that needs to see, what are they doing right… what's working well, applaud them for it and… use that to motivate them to be able to move to the next step.” (nurse practitioner/clinic lead, 4, M)	Weekly buprenorphine E‐Consult VA “Ask the Expert” email service Audit and feedback Community of Practice email highlighting success stories and facility‐wide trends.
Training and educational materials: Training Brochures	“When they do the training through the VA… I feel like that could have been a bit better in the actual prescribing, dosing…equivalents, if you have to switch from one formulation to another…” (nurse practitioner, 30, M) “I really think that education on the provider and staff side is just really a huge piece… have an education program in a way that engages people and their comfort level.” (social worker/pain coordinator, 13, F)	Three‐hour live, case‐based training sessions (virtual or in‐person) for participating community clinics, developed and presented by substance use disorder experts. Buprenorphine quick‐reference guides (Buprenorphine FAQ, OUD and Chronic Pain, Rural Care Model Infographic)

Abbreviations: FAQ, Frequently Asked Questions; OUD, opioid use disorder.

3.2.1. Consultation services

The Multidisciplinary Buprenorphine E‐Consult is a virtual weekly meeting offered to providers within the VA Portland Health Care System, moderated and staffed by institutional leaders and specialists in addiction treatment, pain management, and opioid safety. The team, in collaboration with the referring clinician, discusses patient cases submitted through the electronic health record (E‐Consult) over the prior week. The goal of the consult is to provide mentorship and case discussion to clinicians embedded in diverse settings regarding addiction‐related issues (e.g., appropriate setting of care, treatment options) as well as bridge prescription coverage for patients awaiting more permanent treatment options (e.g., those initiating buprenorphine in the hospital). ⁴⁵

The VA “Ask the Expert” Email Service is a national service that allows any VA employee to send queries on any addiction‐medicine‐related topic to clinical experts in addiction medicine. The email service provides rapid responses to queries—typically within two business days.

3.2.2. Process of audit and feedback

A process of “audit and feedback” was developed to share trends and recognize site improvement. Monthly emails provide data and trends over time in buprenorphine prescribing to targeted community‐based outpatient clinics. Community of practice emails report site‐specific and facility‐wide trends in buprenorphine prescribing across the health care system and highlight positive success stories and inspirational quotes. Sites evidencing significant improvement in primary‐care‐based buprenorphine are recognized in a congratulatory email cc'd by local and national SUD leadership and acknowledged in system‐wide primary care staff meetings.

3.2.3. Training and educational materials

A Buprenorphine Training was developed by the clinical team (KK, KM) tailored to staff within VA community‐based outpatient clinics. Content is case‐based to enhance relevance and engagement, with cases selected to represent common clinical scenarios encountered in primary care. The training begins with an introduction to buprenorphine pharmacology and formulations, addresses medication and OUD stigma, and then uses cases to frame discussion of identifying appropriate buprenorphine candidates, initiating buprenorphine, and common challenges encountered. The training concludes with a discussion of harm reduction principles. Within the training, the team introduces the aforementioned clinical consult services providers can utilize for questions and advice.

Educational Resources address buprenorphine logistics, care processes, and common questions and concerns. Documents incorporate elements of preexisting VA resources, such as materials from the VA Office of Academic Detailing, as well as evidence‐based resources from outside VA, such as resources provided through the Substance Abuse and Mental Health Services Administration. Materials include a Buprenorphine Frequently Asked Questions document (e.g., health benefits of MOUD; language to address stigma) (Appendix A), Indications of Possible Opioid Use Disorder Among Patients Prescribed Long‐Term Opioid Therapy (Appendix B), and a Buprenorphine Care Models Infographic, a visual representation of buprenorphine care models utilized among rural health care systems in our sample (Appendix C).

4. DISCUSSION

In this study, we describe the process of developing an implementation strategy to support buprenorphine prescribing in rural, primary care settings. Our learning health care system research process utilized national administrative data to identify “positive deviant” rural VA health care systems, qualitative interviews to identify key clinical needs and strategies, and a clinical research partnership to inform strategy development. This process allowed us to identify concrete, workable solutions that had been used in other rural VA contexts to overcome common barriers and challenges to buprenorphine prescribing. The successful processes we identified informed the development of our own implementation strategy, composed of an engaging case‐based training, an audit and feedback process, and educational resources, which was adapted to our local context and deployed in partnership with a clinical team. Next steps in our research include formally evaluating the feasibility and acceptability of the strategy, as well its potential impacts.

Echoing prior research, interview participants identified the need for training, ongoing consultation, and educational resources to support providers new to buprenorphine prescribing. ²⁰ , ²¹ , ²⁶ , ²⁹ Within our sample, participants additionally emphasized the importance of clinical mentorship and consultation: to boost confidence, provide encouragement and help move clinicians from intention to action. Future research is needed to identify models of clinical mentorship and/or consult services that that can be formalized (e.g., supported with clinical staffing), replicated, and scaled to ensure broader dissemination and sustainability.

In contrast to prior research, ²⁷ , ³⁰ participants did not identify financial resources or incentives to support prescribing as a key clinical need. This may reflect that VA is an integrated health care delivery and financing system. As such, respondents may be less responsive to financial incentives. ⁴⁶ Nonetheless, cost constraints are an important aspect of health care decision‐making. Costs associated with the implementation strategy presented include clinician time to staff the E‐Consult and conduct the training, which was covered by general salary support (clinical time). Further research is needed to evaluate the costs and benefits of implementing and sustaining buprenorphine prescribing in primary care and the effectiveness of distinct models of care delivery on patient outcomes.

Through this research process, we identified “lessons learned” that may inform others conducting learning health care systems research. First, engagement and collaboration with a clinical partner were crucial. Clinical partners answered questions as they arose, provided context for findings, reviewed and critiqued materials, and helped disseminate research findings. Most importantly, they designed and conducted trainings that were engaging, tailored to local needs, and reflective of the latest science. Second, while access to administrative data is an important strength of learning health care system research, our experience cautions against exclusive reliance on administrative data to identify facility performance. While we utilized administrative data to identify rural health care systems that had improved from 2015 to 2020, given the time lag between the data and interviews, participants in two rural health systems described challenges sustaining primary care buprenorphine, primarily due to staffing turnover. Although all sites were able to speak to the development of their primary care buprenorphine program, in future research, we would utilize both administrative data as well as word‐of‐mouth knowledge and networking to select sites.

This research has limitations. First, we selected sites that were in the process of increasing buprenorphine prescribing in primary care; information gleaned from these sites may not reflect the most successful approaches. Second, while we conducted interviews with staff in diverse clinical roles within each site, additional interviews may have uncovered new themes. Third, although clinical needs and supports requested were consistent across sites, buprenorphine care models varied, and resources and strategies needed to implement these models may likewise have been diverse. Fourth, we did not conduct interviews with patients to ascertain their needs and preferences; future research is needed to understand the patient perspective regarding treatment setting and access. Finally, as the evaluation is ongoing, we do not yet know how this implementation strategy may affect clinical outcomes, nor how this strategy may compare to other efforts active within VA that utilize different approaches.

5. CONCLUSIONS

In a qualitative study that exemplifies learning health care system research, we describe a developmental process that leveraged national administrative data, health care provider interviews, and clinical partnership to develop an implementation strategy to encourage buprenorphine prescribing in rural primary care settings.

FUNDING INFORMATION

This work was supported by the U.S. Department of Veterans Affairs Health Services Research and Development (1IK2HX003007) and resources from the VA Health Services Research and Development‐funded Center to Improve Veteran Involvement in Care (CIVIC) at the VA Portland Health Care System (CIN 13‐404). The content is solely the responsibility of the authors and does not represent the official views of the United States Department of Veterans Affairs or the United States Government.

CONFLICT OF INTEREST STATEMENT

No author reports having any potential conflicts or competing interest with this study.

Supporting information

Appendix S1. Supporting Information.