Strategies to recruit rural primary care providers to implement a medication for opioid use disorder (MOUD) focused integrated care model

Abstract

Background

Access to providers and programs that provide medications for opioid use disorder (MOUD) remains a systemic barrier for patients with opioid use disorder (OUD), particularly if they live in rural areas. The Rural Access to Medication Assisted Treatment (MAT) in Pennsylvania Project (Project RAMP) addressed this problem with a multisystem partnership that recruited, trained, and supported rural primary care providers to provide MOUD and implement an integrated care model (ICM) for patients with OUD. Given the demonstrated efficacy of Project RAMP, this article summarizes our recruitment strategies, including feasibility concerns for further expansion into other regions.

Methods

The approach for recruiting implementation sites included two phases: partner outreach and site identification. Once recruited, the Systems Transformation Framework guided planning and implementation activities. Recruitment and implementation activities were assessed with implementation trackers and evaluated by providers via key informant interviews (KIIs).

Results

Project RAMP recruited 26 primary care practices from 13 counties, including nine health systems and two private practice groups—exceeding the original target of 24 sites. There was a median of 49 days from first contact to project onboarding. A total of 108 primary care practices spanning 22 health systems declined participation. Findings from the KIIs highlighted the value of engaging PCPs by connecting to a shared vision (i.e., improving the quality of patient care) as well as addressing perceived participation barriers (e.g., offering concierge technical assistance to address lack of training or resources).

Conclusion

Findings highlight how successful recruitment activities should leverage the support of health system leadership. Findings also emphasize that aiding recruitment and engagement efforts successfully addressed prescribers’ perceived barriers to providing MOUD as well as facilitating better communication among administrators, PCPs, behavioral health professionals, care managers, and patients.

Plain Language Summary: Opioid use disorder (OUD) is one of the leading causes of preventable illness and death. The standard of care for OUD is the provision of medications for opioid use disorder (MOUD) and the application of an integrative integrated care model (ICM) where behavioral health is blended with specialized medical services. Unfortunately, access to providers and healthcare facilities that provide MOUD or apply an ICM remains a systemic barrier for patients with OUD, particularly if they live in rural areas. Although there is no one-size-fits-all approach to implementing MOUD in primary care, findings from Project The Rural Access to Medication Assisted Treatment (MAT) in Pennsylvania Project (Project RAMP) highlight strategies that may improve future MOUD and ICM implementation efforts in similar rural contexts. Specifically, future efforts to increase MOUD capacity by recruiting new providers should be prepared to leverage health system leadership, address provider barriers via training and expert consultation, and facilitate connections to local behavioral health providers. This approach may be helpful to others recruiting health systems and primary care practices to implement new care models to use MOUD in treating patients with OUD.

Keywords

health services accessibility opiate substitution treatment/methods opioid-related disorders/diagnosis opioid-related disorders/drug therapy Pennsylvania quality of health care/organization & administration rural health services the United States

Introduction and Background

In rural areas of the United States, opioid use disorder (OUD is a leading cause of preventable illness and death (Hedegaard et al., 2020). The current standard of care for OUD is providing medications for OUD (MOUD), including Food and Drug Administration (FDA)-approved medications such as naltrexone, buprenorphine, and methadone (Substance Abuse and Mental Health Services Administration, 2020). Naltrexone can be prescribed by any provider licensed to prescribe medications, but buprenorphine and methadone have more restrictions. Prescribing providers can be certified to prescribe buprenorphine under the Drug Addiction Treatment Act of 2000 with the completion of 8 h (for allopathic and osteopathic physicians) or 24 h (for advanced practice practitioners) of training from a credentialed organization. Upon completion, providers can apply for a waiver on their Drug Enforcement Administration (DEA) license to prescribe buprenorphine for OUD. Recently, the United States exempted the educational requirement for prescribers intending to treat fewer than 30 patients at one time, but providers must still notify the Substance Abuse and Mental Health Services Administration (SAMHSA) and the DEA of their intent to prescribe buprenorphine for OUD. As a Schedule II medication, methadone can only be dispensed at licensed opioid treatment programs.

Despite efforts to increase treatment access, individuals with OUD remain underserved. For example, in 2002, the FDA approved buprenorphine for OUD treatment (U.S. Food and Drug Administration, 2002), and agencies started encouraging primary care providers (PCPs) to screen patients for OUD and provide MOUD in office-based settings (Codell et al., 2020; Gordon et al., 2020; Substance Abuse and Mental Health Services Administration, 2020; Volkow et al., 2014). Yet as of 2018, more than 57% of rural U.S. counties lacked a buprenorphine prescriber (Ghertner, 2019). In nationally representative samples, only 35% of adults with OUD assessed between 2015 and 2017 received any treatment in the past year (Jones & McCance-Katz, 2019). Studies with claims data indicate substantial variability in coverage based on insurance and state of residence, ranging as low as 11% of adults with a suspected OUD based on claims indicators receiving MOUD among commercially insured patients to 40%–80% of comparable Medicaid patients (Donohue et al., 2021; Morgan et al., 2018). This suggests that the regulatory supports and guideline practices have not sufficiently addressed the MOUD access gap.

Unfortunately, patients bear the accumulating consequences of the MOUD access barriers across systems. Provider-perceived barriers include inadequate access to training (e.g., DATA 2000 waiver training), appropriate referral partners for supportive services (e.g., care management and psychosocial counseling), time/resources to adopt new treatments, and finances to support the MOUD services (Andrilla et al., 2018; Bolinski et al., 2019; DeFlavio et al., 2015; Ghertner, 2019; Hutchinson et al., 2014; Mackey et al., 2009; Manuel et al., 2013). Health system barriers to MOUD adoption include a lack of interested/qualified prescribers, insufficient referral and data-sharing infrastructure between treatment providers, as well as inadequate funding for MOUD-related clinical activities (Haffajee et al., 2018; Madras BK et al., 2020). Moreover, the stigma surrounding OUD and MOUD reduces patient and provider participation in MOUD treatment (Crapanzano et al., 2018; van Boekel et al., 2013). These barriers all contribute to ongoing morbidity and mortality from OUD.

The provision of MOUD alone is insufficient for addressing the scope of the opioid overdose death crisis (American Society of Addiction Medicine, 2020). The traditional addiction treatment models where care is siloed into specialty centers limit access and fragment the provision of overall care (Perrone et al., 2019). Drawing from chronic care models which advocate for treating substance use disorders (SUDs) as chronic health conditions within primary care practices (McLellan et al., 2000; McLellan et al., 2013), ICMs provide a structure where patients receive a blend of specialty medical services, such as MOUD, with behavioral health services (Perrone et al., 2019). The blended services within an ICM are coordinated by case managers who support providers as well as coordinate and track patient care across systems (Brackett et al., 2021). Though traditionally focused on depression and anxiety, data also indicate that ICMs for MOUD help lower costs while increasing patient and provider satisfaction (Archer et al., 2012; Brackett et al., 2021; Unützer et al., 2002).

Recognizing the need to increase access to MOUD and integrated care, a multidisciplinary team at the University of Pittsburgh partnered with the Pennsylvania Department of Human Services (DHS) to initiate the Rural Access to Medication Assisted Treatment (MAT¹) in Pennsylvania Project (also known as Project RAMP). Project RAMP was supported by a 4-year Research Demonstration and Dissemination grant led by the DHS and funded by the Agency for Healthcare Research and Quality (AHRQ #: R18-HS025072) that aimed to increase rural access to MOUD by implementing an ICM (Cochran et al., 2019, 2022; Cole et al., 2021). The goal of the Project RAMP ICM was to improve provider capacity to identify, assess, manage, and treat OUD by promoting a team-based approach to patient care and alleviating the PCP from some patient management responsibilities. The Project RAMP ICM provided a structure for connecting PCPs with existing community-based care management, SUD, and mental health treatment resources. PCPs received support to either provide patients with MOUD or then link them to appropriate care management/treatment resources or refer their patients to certified MOUD providers, providers with OUD/SUD treatment experience, or providers able to connect with other ICM services. A prerequisite for effectively implementing Project RAMP was effectively recruiting and engaging providers to participate. Despite repeated calls highlighting the need for improved provider engagement in research and recruitment strategies to facilitate enhanced implementation of evidence-based practices in the community, publications on recruitment strategies remains limited (Department of Health and Human Services; NIH Pragmatic Trials Collaboratory; Nourjah & Kato, 2021; Pinto et al., 2019).

This article has three objectives: (1) describe the recruitment of PCPs and implementation of the MOUD-focused ICM for Project RAMP; (2) describe the results of the Project RAMP recruitment approach in terms of practice engagement and duration of time from “initial contact” to completion of key implementation indicators; and (3) describe the feasibility of this approach to recruit PCPs and develop the MOUD-focused ICMs in other regions, based on provider feedback.

Method

Implementation Framework—Systems Transformational Framework

Project RAMP was guided by the Systems Transformational Framework (Scott & Pringle, 2018) throughout the design, recruitment, pre-implementation, and implementation process. The STF is centered on the concept that all human systems are both complex and adaptive; as such, the process for transforming these systems must involve (1) assessing the functioning of specific system domains, called “levers,” and (2) applying systemic strategies that utilize strong levers within the system to drive change (system transformation) towards the intended outcome. The STF delineates eight levers likely to influence optimal evidence-based practice implementation. The Implementation Team received approximately 6 h of training in the application of the STF.

Recruitment Process

The Project RAMP Implementation Team, based out of the University of Pittsburgh School of Pharmacy Program Evaluation and Research Unit, involved an average of 3.0 full-time equivalents throughout the project. The Implementation Team included persons with public health and data analytic expertise. The Implementation Team worked with the DHS and the Medicaid managed care organizations (MCOs) across the Commonwealth of Pennsylvania to identify potential primary care practices and systems in a set of 23 counties designated as rural by Rural-Urban Continuum Codes (United States Department of Agriculture, 2013) that had OUD prevalence and opioid overdose rates that were above the 2014 national average of 0.9% and 0.0083%, respectively (Han et al., 2015). Ideal candidate practices were those identified as being proximal to a care management entity and/or suitable SUD treatment resources that could support patient care. Care management service providers were located in the Commonwealth's Centers of Excellence (Commonwealth of Pennsylvania Centers of Excellence, 2021) or Single County Authority system, within licensed SUD assessment or treatment agencies, or within the recruited health systems. The Project RAMP goal was to recruit 24 primary care practices across 23 eligible counties over the grant period. These targets were selected to power the primary project analyses (to be reported in a separate manuscript) to detect a difference-in-difference intervention estimate with a 10%–20% increase at .009 desired accuracy (half-width for 95% confidence interval) for 1,000–2,000 intervention beneficiaries and 2,000–3,000 control beneficiaries. All procedures discussed below were considered exempt by the Institutional Review Board of the University of Pittsburgh.

Recruitment for Project RAMP occurred in two phases broken down into four steps, as shown in Figure 1. To determine the duration of the “Recruitment Phase” for each site, we calculated the number of days between initial contact and the onboarding meetings. Consistent with the STF, in the first phase (partner outreach), the Implementation Team engaged leadership from local agencies (governmental and private) providing care management and SUD treatment services and local health system leadership with the objective of developing collaborative relationships that could be leveraged in the recruitment of primary care practices. More specifically, care management and SUD treatment agencies were engaged to develop and implement comprehensive care coordination protocols with the PCPs, thus alleviating expectations on PCPs to provide intensive care management and counseling services in addition to pharmacotherapy. Like PCPs, care management and SUD treatment programs were required to participate in the project's informal health information exchange (HIE), continuous quality improvement (CQI), and performance measurement. Health system leadership was engaged to encourage PCP participation via strong verbal support for MOUD implementation and commitment of system resources to support MOUD provision where possible. Occasionally, initial contact with leadership personnel occurred via facilitated introductions from DHS and the MCO partners. In other cases, the Implementation Team reached out to care management and SUD treatment providers directly via telephone “cold calling.”

Figure 1

Two-Phase Recruitment Process for Rural Access to Medication Assisted Treatment in Pennsylvania Project (Project RAMP)

In the second phase (practice outreach), leadership or practitioners of identified practices were contacted via introduction by care management or SUD treatment programs, health system addiction program leadership, or via “cold calls.” During initial phone calls with prospective practices, the Implementation Team delivered a short, scripted “pitch” about the project and the general need for improving access to MOUD in rural areas. The objective of these calls was to schedule a virtual or in-person meeting with practice leadership and providers, during which they received a formalized recruitment presentation (and an accompanying packet of factsheets) highlighting the anticipated proportion of patients within their practice suffering from OUD/SUD and the general lack of SUD identification through screening and successful referral to services in the United States. Consistent with the STF, the recruitment presentation emphasized the shared vision between the project team and practices as well as the benefits of enhanced coordination with care management service providers and explained aspects of the RAMP approach to supporting MOUD adoption including (1) an individualized implementation process to improve clinical workflow; (2) HIE/CQI activities to support clinical care improvement; (3) provision of concierge technical assistance throughout the implementation process; (4) teleconsultation with physician experts; (5) revenue generation from program-specific activities (i.e., screening, referrals, MOUD-associated activities, care management, etc.); (6) provision of free DATA 2000 waiver training; and (7) provider participation options regarding giving and receiving MOUD treatment referrals, and the option to provide treatment with extended-release naltrexone (also called naltrexone-XR), buprenorphine, or both buprenorphine and naltrexone treatment. Also emphasized during this time was the potential impact that improving provider knowledge about a patient's SUD treatment status and providing referrals to care could have on patient outcomes (Substance Abuse and Mental Health Services Administration, 2020). Occasionally, the Implementation Team would present a modified version of this recruitment pitch (i.e., adjusting presentation delivery of the pitch for different audiences and/or time slots) for a broader audience at network-sponsored recruitment events (rather than individual practices).

Anticipated questions and answers were gathered from project partners and developed into a frequently asked questions (FAQ) document. After the practice or system verbally committed to participation in the project, participants provided insight into how the program could be integrated into their workflow and result in improved patient outcomes. These insights helped the Project RAMP team to engage new provider groups. Occasionally, project partners provided direct support to the Program Evaluation and Research Unit team in conducting recruitment calls and on-site events at healthcare systems, offering testimonials on outcomes supported with data to encourage participation by prospective providers.

Integrated Care Model Pre-Implementation Processes

Once recruited, Project RAMP's Implementation Team designed an implementation approach based on the STF principles and tailored to each site (Scott & Pringle, 2018), then held onboarding meetings with each practice where “MOUD Champions” were identified. In Project RAMP, MOUD Champions promoted MOUD adoption, assisted in coordinating and completing key project activities, as well as facilitated the recruitment of additional practices (Miech et al., 2018). Champion roles included MDs, DOs, nurse practitioners, physician assistants, medical assistants, practice managers, counselors, network directors, care management support specialists, and administrative staff.

The Implementation Team then administered organizational health assessments to each practice to assess their system levels based on the domains of the STF (e.g., leadership, organizational behavior, organizational structure) (Scott & Pringle, 2018). Practices were also asked to provide patient and staff demographics as well as a summary of current SUD screening and treatment processes. These results informed ICM implementation in terms of protocol development, training, data collection, CQI processes, and evaluation. The STF permitted Project RAMP to address issues within the organizations implementing the ICM, create conditions for successful organizational change, and facilitate MOUD adoption. Throughout ICM implementation, the Implementation Team provided on-demand technical assistance using the Lean Principles of teacher/mentor relationship to offer individualized problem-solving support and additional resources (Womack et al., 1991).

Key Informant Interviews (KIIs)

After completing the MOUD implementation, the Project RAMP Evaluation Team conducted semi-structured KIIs with the MOUD providers and clinical staff to assess the impact of the ICM and implementation team on enabling MOUD services. During the interviews, the team collected data to (1) understand the facilitators and barriers to MOUD adoption in rural primary care settings, and (2) receive feedback on Project RAMP. Providers received US$100 for participating. Qualitative data were audio-recorded, transcribed, and imported into NVivo 12 for data management and analysis. Once the data was coded, the Program Evaluation and Research Unit Evaluation Team utilized both inductive and deductive approaches to identify common barriers and facilitators to MOUD adoption at the project sites, with a focus on facilitators that enabled site providers to implement the ICM and adopt MOUD despite known barriers and challenges. A codebook was created with a priori codes based on the interview guide as well as emergent codes grounded in the data (after a process of open coding the transcripts). Two coders (Bridges and Taber), independently coded three transcripts, then met to discuss coding application agreements and disagreements, then refined the codebook accordingly. The current article focuses only on responses related to motivations to engage in Project RAMP. A more detailed description of the methodology and comprehensive KII results related to implementation outcomes will be reported separately (Bridges et al., 2022).

Results

Recruitment Outcomes

A total of 134 primary care practices spanning 17 health systems, four private practice groups, and 23 Pennsylvania counties were contacted for participation in the first recruitment phase. Of the 134 practices contacted, 108 practices declined to participate. Approximately 540 total contact attempts were made with the non-participating practices to engage them in the project. Only 23 practices provided explicit reasons for declining to participate. The remaining 85 practices were unresponsive to outreach attempts and were ultimately classified as “declining” participation (per the Implementation Team's recruitment policy, at least five outreach attempts had to be made before classifying an unresponsive practice as a “decline”). Of the 23 practices that provided a reason for declining to participate, 48% indicated a lack of interest, 26% indicated a lack of health system support for implementation, including a lack of reimbursement structure, and 26% indicated capacity concerns, including competing priorities in managing other chronic conditions and the time needed for training and implementation.

Of the 134 practices that were contacted, 26 practices opted to participate—exceeding the target of 24 and equating to an overall recruitment rate of 19.4%. The final 26 participating practices included nine health systems and two private practice groups spanning 13 counties across Pennsylvania, with the furthest practice located 240 miles from the Implementation Team. To simplify project activity management and coordinate activities of practices within the same practice or health system, some individual PCP practices were combined into one project implementation “site” and are therefore referred to and analyzed as “sites” moving forward. The Implementation Team managed 17 project sites (composed of 26 primary care practices) from recruitment through full implementation.

The time duration of the recruitment stage varied considerably across sites. Specifically, there was a mean of 65 days and a median of 49 days from first contact to project onboarding. The shortest recruitment stage occurred in 12 days, while the longest recruitment stage occurred in 176 days. Table 1 provides descriptive statistics on the characteristics of the sites recruited. For example, at the time of onboarding (i.e., Pre-Implementation), the 17 project sites had 71 practicing PCPs and 148 full-time equivalents (FTEs) clinical staff. Project sites had a range of 2–10 providers, 3–30 clinical staff, and 1–2 MOUD Champions per site. Nine of the project sites had providers that were already DATA waivered. One site prescribed only buprenorphine, two sites prescribed only naltrexone-XR, four sites prescribed both, and two sites prescribed neither medication despite the PCPs having DATA waivers. Of the 17 project sites, eight sites were conducting some form of SUD identification or prevention screening, but inconsistently applying clinically validated or formalized screening tools or protocols.

Table 1

Pre-Implementation Characteristics of Project RAMP Sites Assessed at Onboarding

Site characteristics			Capacity cross sites
Conducting any SUD screening	Prescribing buprenorphine	Prescribing naltrexone	MOUD champions	FTE providers on-site	FTE DATA-waivered providers	Buprenorphine treatment slots^a
n	n	n	n	n	n	N	Range
8	5	6	24	72	15	1,610	30–475

Note. N = 17 sites. MOUD = medication for opioid use disorder; SUD = substance use disorder; SBIRT = Screening, Brief Intervention, Referral to Treatment; FTE = full-time equivalent; N/A = not assessed.

Buprenorphine treatment slots calculated by summing waivered provider limits.

Key Informant Interviews

KIIs were completed with 27 MOUD providers and clinical staff from 12 (of a possible 17) project sites between April 2019 and May 2020 representing a 70.5% response rate. Five project sites declined participation in KIIs due to a lack of interest or time (29.5%). Twenty-five of the interviews were with a single interviewee and one interview had two participants. The average KII was 47 min. The KIIs identified motivations driving providers and staff to engage in Project RAMP as well as ways in which the Project RAMP approach specifically facilitated site engagement and MOUD implementation. Themes and illustrative quotes from providers can all be found in Table 2. The illustrative quotes in Table 2 have also been numbered for ease of reference within the text results below.

Table 2

Themes and Accompanying Provider Quotes From Key Informant Interviews

Motivations to engage in project RAMP	Quote #	Illustrative quotes from providers
Participation dictated by leadership or physicians	1	Doctor came back and said, “We’re doing this.”
Financial incentives to participate	2	I’m also a National Health Service Corps site. And they just had their initiative to try to get more MAT providers. They basically boosted the pay that they were giving …. But you know, with Rural Health Clinic, you get paid at a better rate for Medical Assistance, provided it is a medically-necessary visit.
Financial incentives to participate	3	In addition, we are a Rural Health Clinic, and with the National Health Service Corps, they do offer a loan repayment program. That includes if you are doing a substance use disorder and the MAT. So that also was something that encouraged me to start the process [of providing MAT].
Professional obligation to provide high quality, evidence-based, person-centered care	4	Our vision is to take care of [MAT patients] as a family doctor, or as a primary care clinician or provider. So that we are treating, we are managing you with your substance use disorder as your neuropathy, as well as your depression or what have you … that's really where I see where we see the role of primary care fitting in … maintenance in the context of providing whole-person primary care over the course of time. Longitudinal care.
Professional and civic obligation to fill MOUD service gap in rural communities	5	I felt there was definitely a lack of providers around here that were educated, accountable, and could relate to patients and provide that type of care.
	6	Why am I doing MAT? Because I think it is a hidden population that's not really addressed. There's not a lot of participating providers out there. And there's a lot of stigma to providing that care.
	7	I mean, I’m local-ish. I’m from [town] originally, which was as bad there as it is here. People are dying every day for no good reason due to opioid dependence. If I could do something to help with that, there's a huge need in this area. So, it's something that could definitely be of service.
Past exposure to MOUD training via professional programs and conferences	8	I mean, we learned about it [MOUD] in residency. And I’m only, what, three years, two or three years out of residency. And I feel like it's my responsibility as a primary care provider to provide complete care for my patients. And if there's something I can do to help them prevent deaths, I should do it.
	9	In medical school, I worked in a Methadone clinic for a little while. I rotated through a Methadone clinic. So, I had some experience with treating people back then. …. So, I’ve always had that interest. And then I’ve just reinforced it when in the residency program I was at, and specifically the health center I was at, started their own Medication Assisted Treatment program. So, I was kind of there when that was getting started, which reinforced again. And then. So, I’d always kind of known I wanted to get my waiver to prescribe to help people.
Personal experiences with substance use	10	I brought the idea of starting a MAT program to the administration and they were thrilled because they had asked previous providers if they were interested and no one was interested in getting involved with that. I went and got my waiver and then it was a question of how, where do we start. I probably should tell you right up front, I’m recovering. So, that's where my interest comes from.
Personal experiences with substance use	11	I personally have had family members that have struggled with substance use disorder in the past. And I wished for them to have access to those kinds of providers, because I am from a rural area in Pennsylvania.

Note. MOUD = medication for opioid use disorder; Project RAMP = Rural Access to Medication Assisted Treatment in Pennsylvania Project.

Motivation to Engage in Project RAMP

Many nurses and office managers viewed project engagement and providing MOUD services as part of their job, dictated to them by leadership or physicians. For example, in response to the question, “How did your practice become involved in Project RAMP?” one informant simply stated (Table 2, Quote 1): “Doctor came back and said, ‘We’re doing this.’”

Other providers highlighted the financial incentives associated with being a Rural Health Clinic and National Health Service Corps site, including higher service reimbursement rates and student loan repayment programs for staff (Table 2, Quotes 2 and 3). While providers generally felt that providing MOUD was part of their job, many responses reflected a sense of professional obligation for providing high-quality, evidence-based, person-centered care (Table 2, Quote 4). Because of their rural status, providers felt a heightened sense of professional and personal obligation to fill critical MOUD service gaps in their rural communities (Table 2, Quotes 5–7). While some providers formed these opinions and motivations in part through MOUD exposure in their training programs and at professional conferences (Table 2, Quotes 8 and 9), others were driven to provide MOUD due to personal experiences with substance use disorder (Table 2, Quotes 10 and 11).

Discussion

The goal of Project RAMP was to increase rural primary care patient access to MOUD and integrated care. The present article evaluated the efficacy of the recruitment approach used for Project RAMP to recruit rural PCPs to implement a MOUD-focused integrated care model. We described a two-phase PCP recruitment strategy which included leveraging the support of health system leadership to open doors to network-sponsored recruitment events and provider introductions. Leveraging “early MOUD adopters” within the health system also was important for identifying and gaining commitment in the initial recruitment phase. Results showed that the Project RAMP recruitment approach successfully recruited 26 sites to participate, exceeding the Project's target of 24 sites, and there was a median of 49 days from first contact to project onboarding. Findings from the KIIs highlighted the value of engaging PCPs by connecting to a shared vision (i.e., improving the quality of patient care) as well as addressing participation barriers such as lack of training or resources via concierge technical assistance, training, and expert consultation.

The original recruitment target for Project RAMP was 24 sites; the Implementation Team exceeded that target by successfully recruiting 26 sites from a potential pool of 134 translating to a recruitment rate of 19.4%. The Project RAMP recruitment rate also exceeded rates that have been reported in recent publications on implementation studies in similar settings with a comparable level of intensity in terms of implementation strategies and integration (Ford et al., 2021). For example, a recent publication describing the recruitment rates for a study implementing strategies to improve access to addiction and psychotropic medications reported a response rate of 10.4% in response to recruitment letters (Ford et al., 2021). Other studies reporting higher recruitment rates (e.g., 75%–90%) tend to draw from self-selected participation pools (e.g., sampling from providers who have already agreed to be contacted for project participation) or involve relatively less intensive/complex designs (e.g., brief surveys) (Balio et al., 2022; Varley et al., 2020). Additionally, our findings indicate considerable variability in the time duration of the recruitment stage across sites. These findings may suggest a need for tailored recruitment approaches to successfully recruit and engage rural primary care providers across different sites. Indeed, prior research has indicated that implementing MOUD in rural primary care is more likely to be successful if implementers are attentive to the needs of individual providers, are flexible, tailor recruitment and implementation strategies, and provide ongoing support to meet individualized needs (Nourjah & Kato, 2021; Pinto et al., 2019).

KII findings from the current project also highlight broader systemic problems (e.g., lack of providers in rural counties). While some of the proposed solutions exceed the scope of Project RAMP's purpose, they do offer insight into low-cost recruitment and engagement strategies. For example, advertising existing financial incentives for health systems (e.g., higher service reimbursement rates) and providers (e.g., student loan forgiveness) for providing MOUD in rural counties as a potential benefit to participation is a no-cost way to engage leaders, providers, and clinical staff (Blanco et al., 2020). Furthermore, during the KIIs many providers shared a desire to provide the highest quality care to patients to better serve their communities, fulfill their mission as healthcare providers, and support people like themselves or close family/friends with SUD. This indicates a mechanism of engagement rooted in shared values (e.g., providing compassionate, evidence-based care) and honoring their experiences that go beyond financial incentives or solving logistical hurdles. Such recruitment efforts may prove particularly effective when seeking to identify and engage MOUD Champions early on to facilitate long-term implementation success. Pervading barriers such as internalized stigma that create resistance to any engagement or implementation may prove particularly challenging for projects in the recruitment phase. While the Project RAMP Implementation Team had some success by identifying/employing MOUD Champions, using informal networking opportunities, and having Implementation Team staff trained to handle challenging conversations (e.g., using motivational interviewing techniques, countering stigmatizing beliefs), other strategies such as community-wide anti-stigma campaigns may serve as a fruitful preparatory activity for recruitment (Atkins et al., 2020).

Several limitations should be noted. First, as is common with implementation and research initiatives, the amount of data that was able to be collected in Project RAMP was constrained by limited time, resources, and staffing capacity. Because Project RAMP had site-level benchmarks in terms of recruitment (e.g., recruiting 24 practices) and implementation (e.g., having at least one provider trained in MOUD and one MOUD Champion on site), more detailed staff-level data is not currently available (e.g., MOUD Champion turnover rate). Future projects should consider incorporating multi-level sampling schemes and data tracking for subsequent analysis. Relatedly, only 12/17 sites participated in the KIIs, yielding a response rate of 70.5%; the present findings may be biased to more favorable responses to Project RAMP or not represent the feedback of the remaining who may have had greater time, resources, and staffing constraints. Furthermore, the present article focused on recruitment outcomes. The fact that most practices who opted not to participate did not provide reasons for non-participation is not surprising, however, it does limit our ability to address what the barriers to their participation may have been. Future projects should attempt to garner more detailed non-response data to improve future recruitment efforts. Lastly, this project was conducted before the onset of the COVID-19 pandemic; more recent literature has noted the exceptional impact the pandemic has had on healthcare, public health, and the research systems that exacerbate recruitment challenges (Balio et al., 2022). As workers within these systems continue to confront the direct and residual effects of multiple pandemics (e.g., trauma, burnout), future efforts to integrate evidence-based practices will need to incorporate additional structural and financial supports as well as flexibility around relationship building (e.g., virtual presentations vs. In-person recruiting events) (Balio et al., 2022; Mazurenko et al., 2022).

Replication, Sustainability, and Extensions of Project RAMP

The proposed approach to recruiting practices and implementing MOUD through an ICM could be scaled to other outpatient care settings and systems. As of November 2021, 12 of the 15 sites (80%) that implemented the Project RAMP model and provided MOUD are still providing MOUD using some or most of the system components within the ICM. This continued use of the Project RAMP model is occurring despite the removal of the HIE/CQI process in 2020. The recruitment and implementation strategies for Project RAMP have already been expanded to include 28 additional practices across 12 counties, building the foundation for creating a systematic recruitment and MOUD implementation model to reduce overdose deaths in rural Pennsylvania. Future research focused on characterizing the mechanisms underlying positive organizational changes (e.g., organizational health, provider self-efficacy) (Lewis et al., 2021; Shapiro et al., 2021; Vax et al., 2021) as well as the extent to which organizational or implementation metrics have a demonstrated impact on the patient experience (e.g., individualized care; reduced wait times; appropriate care transitions; wraparound services) is needed. Consistent with recent calls for more patient-centered care (Marchand et al., 2019), future research using these systems to characterize the impact of implementation metrics on patient-reported recovery outcomes (cf., administrative claims data) is needed (Blasi et al., 2021).

Conclusion

There is no one-size-fits-all approach to implementing MOUD in primary care. Engaging providers requires awareness of barriers and facilitators to adoption at the health system, practice, and provider levels. The Project RAMP recruitment strategy was successful, in part, by establishing a strong commitment from health system leadership and individual practices as well as leveraging successfully recruited providers/sites to recruit additional sites. The ICM aided recruitment and engagement efforts by successfully addressing prescribers’ perceived barriers to providing MOUD (e.g., paying providers fees for obtaining DATA 2000 waivers) as well as facilitating better communication among administrators, PCPs, behavioral health professionals, care managers, and patients. Future initiatives to increase MOUD capacity in rural counties should be prepared to (1) leverage the support of health system leadership in recruiting new PCPs; (2) articulate how they plan to address provider barriers via concierge technical assistance, training, and expert consultation; and (3) facilitate connections with local behavioral health providers and develop an ICM for MOUD via case management.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Agency for Healthcare Research and Quality [grant number R18-HS025072].

ORCID iD

Renee M. Cloutier

Notes

References

American Society of Addiction Medicine. (2020). The ASAM National Practice Guideline for the Treatment of Opioid Use Disorder. 2020 Focused Update.

Andrilla

C. H. A.

Coulthard

Patterson

D. G.

(2018). Prescribing practices of rural physicians waivered to prescribe buprenorphine. American Journal of Preventive Medicine, 54(6), S208–S214. https://doi.org/10.1016/j.amepre.2018.02.006

Archer

Bower

Gilbody

Lovell

Richards

Gask

Dickens

Coventry

(2012). Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews, 10, https://doi.org/10.1002/14651858.CD006525.pub2

Atkins

Dopp

A. L.

Temaner

E. B.

(2020). Combatting the Stigma of Addiction—The Need for a Comprehensive Health System Approach. NAM Perspectives. https://doi.org/10.31478/202011d

Balio

C. P.

Riley

S. R.

Grammer

Weathington

Barclay

Jonas

D. E.

(2022). Barriers to recruiting primary care practices for implementation research during COVID-19: A qualitative study of practice coaches from the Stop Unhealthy (STUN) Alcohol Use Now trial [Article]. Sage Publications. https://doi.org/http://localhost/files/7h14b069k

Blanco

Ali

M. M.

Beswick

Drexler

Hoffman

Jones

C. M.

Wiley

T. R. A.

Coukell

(2020). The American Opioid Epidemic in Special Populations: Five Examples [Discussion Paper]. NAM Perspectives, National Academy of Medicine. https://doi.org/10.31478/202010b

Blasi

P. R.

Mettert

K. D.

Coleman

Lewis

Wagner

Coghill

M. N.

Dang

Richards

J. E.

(2021). Transitioning patients from outpatient mental health services to primary care: A rapid literature review. Implementation Research and Practice, 2, 26334895211041294. https://doi.org/10.1177/26334895211041294

Bolinski

Ellis

Zahnd

W. E.

Walters

McLuckie

Schneider

Jenkins

W. D.

(2019). Social norms associated with nonmedical opioid use in rural communities: a systematic review. Translational Behavioral Medicine. https://doi.org/10.1093/tbm/ibz129

Brackett

C. D.

Duncan

Wagner

J. F.

Fineberg

Kraft

(2021). Multidisciplinary treatment of opioid use disorder in primary care using the collaborative care model. Substance Abuse, 1–5. https://doi.org/10.1080/08897077.2021.1932698

10.

Bridges

Taber

Foulds

Bear

T. M.

Cloutier

R. M.

McDonough

Lomauro

Gordon

Cochran

Donohue

Adair

DiDomenico

Pringle

J. L.

Gellad

Kelley

Cole

E. S.

(2022). Medications for Opioid Use Disorder in Rural Primary Care Practices: Patient and Provider Experiences [Manuscript Submitted for Publication, December, 2022].

11.

Cochran

Cole

E. S.

Warwick

Donohue

J. M.

Gordon

A. J.

Gellad

W. F.

Bear

Kelley

DiDomenico

Pringle

. (2019). Rural access to MAT in Pennsylvania (RAMP): A hybrid implementation study protocol for medication assisted treatment adoption among rural primary care providers. Addiction Science & Clinical Practice, 14, 25. https://doi.org/10.1186/s13722-019-0154-4

12.

Cochran

Cole

E. S.

Sharbaugh

Nagy

Gordon

A. J.

Gellad

W. F.

Pringle

Bear

Warwick

Drake

Chang

C.-C. H.

DiDomenico

Kelley

Donohue

. (2022). Provider and patient-panel characteristics associated with initial adoption and sustained prescribing of medication for opioid use disorder. Journal of Addiction Medicine, 16(2), e87. https://doi.org/10.1097/ADM.0000000000000859

13.

Codell

Kelley

A. T.

Jones

A. L.

Dungan

M. T.

Valentino

Holtey

A. I.

Knight

T. J.

Butz

Gallop

Erickson

Patton

Hyte-Richins

L. J.

Rollins

B. Z.

Gordon

A. J.

(2020). Aims, development, and early results of an interdisciplinary primary care initiative to address patient vulnerabilities. American Journal of Drug and Alcohol Abuse, 1–10. https://doi.org/10.1080/00952990.2020.1832507

14.

Commonwealth of Pennsylvania Centers of Excellence (2021). Retrieved November 18 from www.dhs.pa.gov/Services/Assistance/Pages/Centers-of-Excellence.aspx.

15.

Cole

E. S.

DiDomenico

Green

Heil

S. K. R.

Hilliard

Mossburg

S. E.

Sussman

A. L.

Warwick

Westfall

J. M.

Zittleman

Salvador

J. G

. (2021). The who, the what, and the how: A description of strategies and lessons learned to expand access to medications for opioid use disorder in rural America. Substance Abuse, 42(2), 123–129. https://doi.org/10.1080/08897077.2021.1891492

16.

Crapanzano

K. A.

Hammarlund

Ahmad

Hunsinger

Kullar

(2018). The association between perceived stigma and substance use disorder treatment outcomes: A review. Substance Abuse and Rehabilitation, 10, 1–12. https://doi.org/10.2147/SAR.S183252

17.

DeFlavio

J. R.

Rolin

S. A.

Nordstrom

B. R.

Kazal

L. A. J.

(2015). Analysis of barriers to adoption of buprenorphine maintenance therapy by family physicians. Rural and Remote Health, 15, 3019. https://doi.org/10.22605/RRH3019

18.

Department of Health and Human Services. HEAL Initiative: Research to Foster an Opioid Use Disorder Treatment System Patients Can Count On (RM1 - Clinical Trial Optional) Retrieved August 31 from https://grants.nih.gov/grants/guide/rfa-files/RFA-DA-23-046.html.

19.

Donohue

J. M.

Jarlenski

M. P.

Kim

J. Y.

Tang

Ahrens

Allen

Austin

Barnes

A. J.

Burns

Chang

C. H.

Clark

Cole

Crane

Cunningham

Idala

Junker

Lanier

Mauk

McDuffie

M. J.

Mohamoud

Pauly

Sheets

Talbert

Zivin

Gordon

A. J.

Kennedy

(2021). Use of medications for treatment of opioid use disorder among US Medicaid enrollees in 11 states, 2014–2018. JAMA, 326(2), 154–164. https://doi.org/10.1001/jama.2021.7374

20.

Ford

J. H.,

2nd Kaur

Rao

Gilson

Bolt

D. M.

Garneau

H. C.

Saldana

McGovern

M. P.

(2021). Improving medication access within integrated treatment for individuals with co-occurring disorders in substance use treatment agencies. Implement Res Pract, 2, https://doi.org/10.1177/26334895211033659

21.

Ghertner

(2019). US Trends in the supply of providers with a waiver to prescribe buprenorphine for opioid use disorder in 2016 and 2018. Drug and Alcohol Dependence, 204, 107527. https://doi.org/10.1016/j.drugalcdep.2019.06.029

22.

Gordon

A. J.

Drexler

Hawkins

E. J.

Burden

Codell

N. K.

Mhatre-Owens

Dungan

M. T.

Hagedorn

(2020). Stepped Care for Opioid Use Disorder Train the Trainer (SCOUTT) initiative: Expanding access to medication treatment for opioid use disorder within veterans’ health administration facilities. Substance Abuse, 41(3), 275–282. https://doi.org/10.1080/08897077.2020.1787299

23.

Haffajee

R. L.

Bohnert

A. S. B.

Lagisetty

P. A.

(2018, 01/06). Policy pathways to address provider workforce barriers to buprenorphine treatment. American Journal of Preventive Medicine, 54(6, Supplement 3), S230–S242. https://doi.org/10.1016/j.amepre.2017.12.022

24.

Han

Compton

W. M.

Jones

C. M.

Cai

(2015). Nonmedical prescription opioid use and use disorders among adults aged 18 through 64 years in the United States, 2003–2013. JAMA, 314(14), 1468–1478. https://doi.org/10.1001/jama.2015.11859

25.

Hedegaard

Miniño

Warner

(2020). Drug overdose deaths in the United States, 1999–2018. NCHS Data Brief(no 356).

26.

Hutchinson

Catlin

Andrilla

C. H.

Baldwin

L. M.

Rosenblatt

R. A.

(2014). Barriers to primary care physicians prescribing buprenorphine. Annals of Family Medicine, 12(2), 128–133. https://doi.org/10.1370/afm.1595

27.

Jones

C. M.

McCance-Katz

E. F.

(2019, 2019/04/01/). Co-occurring substance use and mental disorders among adults with opioid use disorder. Drug and Alcohol Dependence, 197, 78–82. https://doi.org/10.1016/j.drugalcdep.2018.12.030

28.

Lewis

C. C.

Mettert

Lyon

A. R.

(2021). Determining the influence of intervention characteristics on implementation success requires reliable and valid measures: Results from a systematic review. Implementation Research and Practice, 2, 2633489521994197. https://doi.org/10.1177/2633489521994197

29.

Mackey

Veazie

Anderson

Bourne

Peterson

(2009). Evidence Brief: Barriers and Facilitators to Use of Medications for Opioid Use Disorder. Evidence Synthesis Program, VA ESP Project #09-009.

30.

Madras BK

A. N.

Wen

Sharfstein

, & the Prevention, Treatment, and Recovery Working Group of the Action Collaborative on Countering the U.S. Opioid Epidemic (2020). Improving Access to Evidence-Based Medical Treatment for Opioid Use Disorder: Strategies to Address Key Barriers Within the Treatment System. NAM Perspectives. Discussion Paper, Washington, DC.

31.

Manuel

Newville

Larios

Sorensen

(2013). Confidentiality protections versus collaborative care in the treatment of substance use disorders. Addiction Science & Clinical Practice, 8(1), 13. https://doi.org/10.1186/1940-0640-8-13

32.

Marchand

Beaumont

Westfall

MacDonald

Harrison

Marsh

D. C.

Schechter

M. T.

Oviedo-Joekes

(2019). Conceptualizing patient-centered care for substance use disorder treatment: Findings from a systematic scoping review. Substance Abuse Treatment, Prevention, and Policy, 14(1), 37. https://doi.org/10.1186/s13011-019-0227-0

33.

Mazurenko

Sanner

Apathy

N. C.

Mamlin

B. W.

Menachemi

Adams

M. C. B.

Hurley

R. W.

Erazo

S. F.

Harle

C. A.

(2022). Evaluation of electronic recruitment efforts of primary care providers as research subjects during the COVID-19 pandemic. BMC Prim Care, 23(1), 95. https://doi.org/10.1186/s12875-022-01705-y

34.

McLellan

A. T.

Lewis

D. C.

O'Brien

C. P.

Kleber

H. D.

(2000). Drug dependence, a chronic medical illness: Implications for treatment, insurance, and outcomes evaluation. JAMA, 284(13), 1689–1695. https://doi.org/10.1001/jama.284.13.1689

35.

McLellan

A. T.

Starrels

J. L.

Tai

Gordon

A. J.

Brown

Ghitza

Gourevitch

Stein

Oros

Horton

(2013). Can substance use disorders be managed using the chronic care model? Review and recommendations from a NIDA consensus group. Public Health Reviews, 35(2), 8. https://doi.org/10.1007/BF03391707

36.

Miech

E. J.

Rattray

N. A.

Flanagan

M. E.

Damschroder

Schmid

A. A.

Damush

T. M.

(2018). Inside help: An integrative review of champions in healthcare-related implementation. SAGE Open Medicine, 6, 2050312118773261–2050312118773261. https://doi.org/10.1177/2050312118773261

37.

Morgan

J. R.

Schackman

B. R.

Leff

J. A.

Linas

B. P.

Walley

A. Y.

(2018). Injectable naltrexone, oral naltrexone, and buprenorphine utilization and discontinuation among individuals treated for opioid use disorder in the United States commercially insured population. Journal of Substance Abuse Treatment, 85, 90–96. https://doi.org/10.1016/j.jsat.2017.07.001

38.

National Academies of Sciences Engineering and Medicine (2019). Medications for Opioid Use Disorder Save Lives .

39.

NIH Pragmatic Trials Collaboratory. Building Partnerships and Teams to Ensure a Successful Trial. Retrieved August 31 from https://rethinkingclinicaltrials.org/chapters/engaging-stakeholders/engaging-stakeholders-and-building-partnerships-to-ensure-a-successful-trial.

40.

Nourjah

Kato

(2021). “One size does not fit all” and other lessons learned from grants for implementation of the AHRQ medication assisted treatment for opioid use disorder in rural primary care. Substance Abuse, 42(2), 136–139. https://doi.org/10.1080/08897077.2021.1891600

41.

Perrone

Weiner

S. G.

Nelson

L. S.

(2019). Stewarding recovery from the opioid crisis through health system initiatives. The Western Journal of Emergency Medicine, 20(2), 198–202. https://doi.org/10.5811/westjem.2018.11.39013

42.

Pinto

R. M.

Spector

A. Y.

Rahman

(2019). Nurturing practitioner-researcher partnerships to improve adoption and delivery of research-based social and public health services worldwide. International Journal of Environmental Research and Public Health, 16(5). https://doi.org/10.3390/ijerph16050862

43.

Scott

K. A.

Pringle

(2018). The power of the frame: Systems transformation framework for health care leaders. Nursing Administration Quarterly, 42(1), 4–14. https://doi.org/10.1097/NAQ.0000000000000261

44.

Shapiro

C. J.

Watson MacDonell

Moran

(2021). Provider self-efficacy in delivering evidence-based psychosocial interventions: A scoping review. Implementation Research and Practice, 2, 2633489520988258. https://doi.org/10.1177/2633489520988258

45.

Substance Abuse and Mental Health Services Administration (2020). Treatment Improvement Protocol: TIP 63. Medications for Opioid Use Disorder for Healthcare and Addiction Professionals, Policymakers, Patients, and Families.

46.

United States Department of Agriculture (2013). Rural-Urban Continuum Codes . www.ers.usda.gov/data-products/rural-urban-continuum-codes.

47.

Unützer

Katon

Callahan

C. M.

Williams Jr

J. W.

Hunkeler

Harpole

Hoffing

Della Penna

R. D.

Noël

P. H.

Lin

E. H.

(2002). Collaborative care management of late-life depression in the primary care setting: A randomized controlled trial. JAMA, 288(22), 2836–2845. https://doi.org/10.1001/jama.288.22.2836

48.

U.S. Food and Drug Administration (2002). FPL for approved NDA 20-732 and NDA 20-723 .

49.

van Boekel

L. C.

Brouwers

E. P. M.

van Weeghel

Garretsen

H. F. L.

(2013). Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: Systematic review. Drug and Alcohol Dependence, 131(1), 23–35. https://doi.org/10.1016/j.drugalcdep.2013.02.018

50.

Varley

A. L.

Lappan

Jackson

Goodin

B. R.

Cherrington

A. L.

Copes

Hendricks

P. S.

(2020). Understanding barriers and facilitators to the uptake of best practices for the treatment of co-occurring chronic pain and opioid use disorder. Journal of Dual Diagnosis, 16(2), 239–249. https://doi.org/10.1080/15504263.2019.1675920

51.

Vax

Gidugu

Farkas

Drainoni

M.-L.

(2021). Ready to roll: Strategies and actions to enhance organizational readiness for implementation in community mental health. Implementation Research and Practice, 2, 2633489520988254. https://doi.org/10.1177/2633489520988254

52.

Volkow

N. D.

Frieden

T. R.

Hyde

P. S.

Cha

S. S.

(2014). Medication-assisted therapies — Tackling the opioid-overdose epidemic. New England Journal of Medicine, 370(22), 2063–2066. https://doi.org/10.1056/NEJMp1402780

53.

Womack

Jones

Roos

(1991). The Machine That Changed The World: The Story of Lean Production.