Increasing Capacity for Treatment of Opioid Use Disorder in Rural Primary Care Practices

INTRODUCTION

The United States has an epidemic of opioid use disorder (OUD). Over 4 million Americans report using prescription pain medicine for nonmedical reasons.¹ First-time heroin users increased from 90,000 in 2006 to 170,000 in 2016.² The use of illicit fentanyl analogs accounted for 12 times more overdose deaths in 2019 than in 2013.³ More Americans die each year from drug overdoses than from motor vehicle accidents.⁴ Beyond mortality, opioid dependence and OUD affects the quality of life of the people with the condition and their families and communities.⁵ Evidence-based guidelines support use of the buprenorphine (a partial opioid antagonist) to treat opioid dependence and OUD in primary care settings.⁶ Putting the guidelines into practice has been slowed by complex definitions, misconceptions about the patients that need treatment, and inadequately trained staff.^7,8 These factors contribute to a serious treatment gap for OUD; it is estimated that only 20% of youth and adults with OUD receive treatment.⁹ Many reported barriers to treatment are magnified in rural communities and primary care.^10,11

Clinicians, practices, and community members affiliated with the High Plains Research Network (HPRN) and the Colorado Research Network (CaReNet) identified opioid dependence as a primary concern in their communities. As the research team visited practices, clinicians and practice staff described their concerns about over-prescribing, drug diversion, potential for addiction, and the severe lack of local treatment. Colorado data supported these concerns, as 7 counties in the HPRN and southern CaReNet regions had drug overdose rates greater than 20 per 100,000; some of the highest rates in the nation.⁴

To help change the trajectory of this epidemic in rural Colorado, the HPRN and CaReNet developed and implemented a multilevel program called Implementing Technology and Medication Assisted Treatment Team Training in Rural Colorado (IT MATTTRs) for medical treatment of OUD with buprenorphine. Efforts to increase treatment for opioid dependence and OUD with buprenorphine included support for clinicians to obtain the Drug Enforcement Agency (DEA) waiver required to prescribe buprenorphine. Studies suggest that simply having a waivered prescriber, however, may be insufficient for delivery of treatment.¹² A frequently reported barrier is an unprepared practice and staff. To address this barrier, primary care implementation models aim to engage the practice team, particularly at practices with a waivered clinician or one actively seeking the waiver.^13,14 Programs using tele-education models with interactive sessions, access to a expert clinician who prescribes buprenorphine for OUD, and case-based learning, are challenged by inconsistent or low attendance.^15,16

Grounded in participatory research, IT MATTTRs builds on these approaches to improve awareness and use of buprenorphine for treatment of OUD. Community-focused interventions were developed and implemented to create conversations about OUD while increasing knowledge of OUD, treatment with buprenorphine, and treatment-seeking behaviors.¹⁷ Clinicians were offered the required DEA waiver training at no cost. The program also delivered a training for practice teams about OUD treatment with buprenorphine to increase patients ability to receive treatment locally. The training was based on advanced primary care and patient-centered medical home literature that promotes a team approach to patient care. The training was delivered to practices with and without a waivered clinician and to practices interested in providing treatment. This manuscript describes the impact of the IT MATTTRs intervention on the implementation and delivery of treatment of OUD with buprenorphine in rural primary care practices.

METHODS

This study was conducted in the HPRN and CaReNet practice-based research networks. The High Plains Research Network consists of 53 primary care practices and communities in the 16 counties of eastern Colorado. The Colorado Research Network consists of 45 practices, most of which care for underserved patients at federally qualified health centers or other practices serving patients with fewer resources. Both networks are housed in the University of Colorado Department of Family Medicine and are members of the State Network of Colorado Ambulatory Practices and Partners.

The IT MATTTRs Practice Team Training was developed by the research team, local medication-assisted treatment experts, community members, and professional practice facilitators.¹⁸ Based on the American Society of Addiction Medicine and Center for Disease Control and Prevention guidelines, the training covered the epidemiology of OUD, pharmacology of buprenorphine, neurobiology of addiction, and detailed treatment steps. The training included facilitated discussion among office staff, nurses, medical assistants, clinicians, and billing and coding staff to encourage the identification of existing resources on which to build and guide implementation. Practices received training via the Shared Onsight Knowledge Dissemination Team Training model, typically delivered in 4 sessions during the lunch hour at the practice, or an Extension for Community Health Outcomes model.

The study aimed to recruit 40 practices based on power calculations to detect pre-post changes. Seventy-nine primary care practices (all HPRN practices and the rural CaReNet practices) in a 24-county region were identified from network rosters and were invited to participate. Enrollment closed when the recruitment goal was reached. Training was delivered over a 20-month period beginning in September 2017. Study protocols were approved by the Colorado Multiple Institutional Review Board.

RESULTS

Forty-two practices enrolled and completed the IT MATTTRs Practice Team Training. Table 1 describes the participating practices.

Participating practices significantly increased the treatment components implemented (Table 2). Practices reported an average 4.7 items in place at baseline compared with 13.0 at 12-month follow-up (F[2,56] = 31.17, P <.001). Practices that had a DEA-waivered buprenorphine prescriber at baseline or acquired one during the study reported a significantly greater increase in treatment components implemented than practices without a waivered prescriber (F[2,56] = 7.77, P = .001). However, the average number of items still increased significantly in practices without a waivered prescriber (baseline 4.1 items; follow-up 8.4 items, P <.001) compared with practices having a waivered prescriber (baseline 4.9 item; follow-up 16 items, P <.001). Items that changed significantly from baseline to follow-up included having a prescriber with waiver to prescribe buprenorphine (16% vs 56%, P = .001), screening (44% vs 82%, P <.01), and having a patient treatment agreement (16% vs 51%, P <.01). At final follow-up, 74% of participating practices reported either providing treatment with buprenorphine or referring their patients for treatment compared with 19% at baseline.

As reported by practices, the overall number of patients prescribed buprenorphine increased significantly (F[1,403] df=6.80, P = .01). New OUD treatment inductions did not increase significantly (P = .20), however, patients in treatment for greater than 6 months did (F[1,403] df = 30.56, P <.001) (Figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Patients prescribed buprenorphine and those in treatment with buprenorphine for 6 months or more reported by IT MATTTRs study practices.

Prescription Drug Monitoring Program data show significant increases in patients per year with a prescription for buprenorphine with a zip code in the 24-county study region (time: overall F[4,1401] df = 209.71, P <.001) and with a prescription for buprenorphine from a local prescriber in the study region (time: overall F[4,1423] df = 652.97, P <.001) (Figure 2). Additionally, patients who obtained their medication from a pharmacy in the study region increased significantly (time: overall F[4,1401] df = 655.84, P <.001). The percent increase in patients with a buprenorphine prescription for the study region was 87% compared with 65% in the rest of the state. Poisson regression indicated that the increase in number of patients with a buprenorphine prescription was significantly greater in the study region from 2015 through 2019 compared with the rest of the state (year x region coef [SE]: 0.0545 [0.0137], Wald χ² = 15.73, P <.001).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Patients prescribed buprenorphine by patient and provider location in the IT MATTTRs study region from 2015-2019.

DISCUSSION

The IT MATTTRs program was associated with an increase of opioid dependence and OUD treatment with buprenorphine in rural Colorado primary care practices. Several factors may have contributed to this increase. First, the IT MATTTRs community-level interventions were associated with community members’ awareness of treatment for OUD in primary care.¹⁷ Second, the IT MATTTRs Practice Team Training reached 441 team members. Across roles, participants reported high levels of satisfaction and significantly higher self-rated ability to deliver treatment components after the training.¹⁸ Third, practices implemented many of the processes and components necessary for delivering treatment.

Across participating practices, treatment components implemented significantly increased. While increases were greater in practices with a waivered clinician, changes were still significant in those without. Practices with protocols to screen, protocols to refer patients to a buprenorphine prescriber, and at least 1 patient referred to a prescriber increased significantly. For some practices, engaging in OUD treatment may begin in partnership with another practice or behavioral health center with a waivered prescriber. While the team training encourages progression toward providing the full spectrum of care, including induction, the fostering of partnerships with this type of intervention may serve as a strategy to increase access to treatment in rural communities.

Practices reported significant increases in patients receiving medication after the intervention and receiving treatment with buprenorphine for at least 6 months. An important message and discussion topic in the team training is that treatment with buprenorphine for OUD requires more than a few months. These results suggest the positive impact of the team training on practices’ expectations and, consequently, patients’ continued treatment engagement. Analysis of practice data did not show a significant increase in new medication-assisted treatment inductions. One explanation could be early pent-up demand in small communities leveling off or decreasing as eligible and interested patients receive care.

Prescription Drug Monitoring Program data indicated a significant increase in people with a prescription for buprenorphine in the study region. People living in rural communities may access medical care outside their local region.²⁴ Since training the practice teams took longer than the community-focused interventions, some patients may have accessed care outside the 24-county study region. Our results also show a significant increase in patients who received their buprenorphine prescriptions locally, based on prescriber location. This outcome suggests that IT MATTTRs, with activated community members and practice teams, was associated with both an overall and a local increase in treatment of OUD with buprenorphine.

Several other statewide and national programs and resources related to OUD prevention, treatment, and policy were ongoing during this study. The combined impact of these likely contributed to a substantial increase in the delivery of treatment in the state as a whole. Our results, however, show a greater increase in patients receiving treatment with buprenorphine in the study region compared with the rest of the state, suggesting an independent effect on improving access to buprenorphine treatment in primary care.

To further support efforts to increase access to treatment, the training is available to primary care practices, hospitals, health systems, and other disciplines and has been delivered to health professionals in Colorado, Montana, California, and North Carolina.

Limitations include missing practice-level data. In some cases, the requested baseline data was not received. When the practice’s training start date could not be altered, we proceeded to avoid losing the practice entirely. However, results from sensitivity analyses with complete data only were similar, supporting the robustness of study findings. The association of individual implementation checklist components and treatment of OUD with buprenorphine is unknown, and analysis is beyond the scope of this paper. Also, approximately two-thirds of participating practices were able to provide data on treatment delivery. The results of this analysis, however, align with the change in the delivery and receipt of treatment demonstrated by PDMP data. This study was not designed to measure quality of care. Last, we were unable to conduct exploratory analysis of patient-level outcomes data as planned to describe the impact of treatment on quality of life. These data were to be obtained from a web-based opioid and patient engagement and monitoring system offered for free to participating practices. Due to practices’ very low uptake of this system, resulting data were too limited for analysis.

CONCLUSION

Improved identification and treatment of patients with OUD in rural primary care practices is critical and timely. A multilevel, community- and practice-focused intervention to implement medication-assisted treatment for OUD effectively increased delivery of treatment in rural Colorado. IT MATTTRs addresses elements beyond the clinician waiver training to make the implementation of treatment of OUD with buprenorphine in practice feasible.