Improving Management of Opioid Use Disorder Training With Novel Resident Co-PCP Model

Medications for opioid use disorder (MOUD), including office-based provision of buprenorphine, is an evidenced-based strategy for managing opioid use disorder (OUD) in primary care.^1-3 It has been demonstrated to improve mortality,¹ retention in treatment,^1,3-6 and quality of life.⁷ Despite the recommendations from multiple organizations,^1,8-10 less than 30% of patients receive this care, linked to deficits in health care training.¹¹

Tong et al assessed buprenorphine provision by early-career family physicians in 2016 and found that only 10% of recent residency graduates reported preparedness to provide buprenorphine treatment. It has separately been shown that dedicated training for health care professionals can improve knowledge, skills, attitudes, and confidence in the care of patients with substance use disorders (SUD).^{12, 16-22}

Residency training appears to be a key opportunity to improve the competency of the future primary care workforce to provide evidence-based treatment for OUD. We sought to identify and remove barriers to integrating residents into office-based MOUD care teams by implementing a novel model of care.

This study was performed at a 4-year family medicine residency in the Pacific Northwest. Residents train at a tertiary care center that includes a 2-week addictions medicine rotation and Drug Addiction Treatment Act of 2000 (DATA) waiver training. The co-primary care provider (co-PCP) model was implemented at one of the five residency’s sites that trains eight of the 50 residents in the program. This site is a designated rural health center (RHC) that had an established MOUD treatment program involving care management, behavioral health, and medical prescriber visits.²³ At the time of this project, eight of 11 faculty preceptors had DATA waiver training. The project received an institutional review board designation of “Not Human Research.”

We performed an appraisal of the clinic’s faculty and resident MOUD patient panels using an electronic health record. No patient information was recorded.

Between November 2018 and June 2019, we implemented a model to assign both a resident and faculty physician to patients receiving MOUD. This echoed an established practice model for prenatal care in place to ensure supervision, continuity of care, and consistent provider access. The model also had the goal of improving patient experience by allowing patients to have two consistent providers with whom they could build trust, given that residents are frequently on rotations outside of their continuity practice site. Finally, this program sought to improve preceptor access, so that residents could call the co-PCP to discuss a plan if the faculty on site did not possess a DATA waiver. The ultimate goal was to increase the residents’ involvement in the care of patients utilizing MOUD, as well as the likelihood that they would practice MOUD after residency graduation.

We aimed to have each resident be assigned to a minimum of three patients receiving MOUD due to a prior program survey showing that our residents caring for only one or two patients were not comfortable prescribing MOUD after graduation.

In order to achieve this, a dedicated MOUD front office staff assisted in assignments and scheduling to alternate between the resident and faculty physician. Researchers utilized an electronic health record tracking system to note who the primary care provider was and also who the resident and faculty MOUD co-PCPs were, important in cases where primary care providers were not yet offering MOUD care. For existing patients on MOUD, faculty providers were asked which patients they believed would be receptive to a resident co-PCP; that list was shared with the dedicated staff, and patients were then assigned to a resident as well.

Preimplementation, eight of 153 MOUD patients at the RHC had a resident provider for MOUD care. This increased to 26 patients after implementation. Preimplementation, six out of eight (75%) residents had no MOUD patients. Within eight months, every resident had at least three MOUD patients, with a range of two to five patients per resident.

With the need for increased access to treatment for OUD, the ideal positioning of family physicians to provide that care, and the current lack of adequate resident training in MOUD nationally, residency experiences that improve MOUD training are needed. This co-PCP model is a novel way to address the systems barriers, increase resident competency and comfort, and account for academic faculty who may not be skilled in MOUD.

While other data were not explicitly collected, during the time of this study the number of faculty and residents informally reporting confidence in MOUD also expanded. For instance, one additional preceptor completed his DATA training, and the remaining two who are not trained were willing to precept and discuss this care with support from other faculty. This may be secondary to gaining confidence through exposure to residents who more frequently had these visits.

Limitations of this study include the small cohort of the clinic and limited time frame of the project. While the co-PCP model has continued, grant funding and staffing shortages have hindered the success of assigning a resident to a growing number of MOUD patients, and some providers and patients chose to continue to have their one-to-one relationship. This is important as others consider adopting such a model. Finally, this study was conducted at an institution with significant resident interest in the care of individuals with OUD. It is possible that other programs would have variable success if lower interest existed within their resident cohorts.

While we did not scope this study to assess the long-term impact of the co-PCP model, there is at least initial support that it may increase resident exposure to MOUD care that could be duplicated in other practices to assess for generalizability and wider effects. Future studies of entire programs nationwide should be considered, such as graduate surveys to assess the downstream effects of the implementation of this new model on future practice patterns. However, this is one practical, straightforward solution that other residency programs could implement with the goal of increasing the number of physicians providing MOUD after residency and expanding overall access to MOUD for patients in need.