Chronic pain is a common, serious illness often accompanied by psychological comorbidities including anxiety and depression (1, 2), decreased quality of life (1), reduced participation in the workforce (2), and strain on interpersonal relationships (2). It affects about 55 million people in the United States (3) and costs $635 billion annually in medical treatment and lost productivity (4). The consequences of chronic pain may be further complicated by environmental factors such as stress and lack of social support (5).

Evidence-based practices (EBPs) to treat chronic pain include pharmacologic therapies [e.g., antidepressants (6), amitriptyline, etc.], non-pharmacologic approaches such as physical therapy (7–9) and cognitive behavioral therapy (CBT) (10–12), and behavioral approaches such as self-management and cognitive behavioral therapy, yoga, and physical therapy (13). However, patients often do not receive evidence-based chronic pain management (14–16). A prime example is the treatment for chronic low back pain. A recent systematic review reported that chronic low back pain prevalence rises linearly from the third decade of life on with rates of 4.2% in individuals between 24 and 39 years old and 19.6% in those ages 20 and 59 (17), with an estimated 2.06 million cases per year (18). A 2018 series published by The Lancet examining current and best practices for treatment showed that low back pain results in 2.7 million emergency department visits each year, despite best practices for managing chronic pain in primary care and avoiding emergency care (19). Further, surgical procedures are often used as a strategy to manage chronic low back pain, which again contradicts guidance advising the reliance on noninvasive treatment strategies (19). One study found that only 12% of people with chronic low back pain have received psychological support and only 8.4% received CBT (19).

Primary care practitioners (PCPs; i.e., physicians, nurse practitioners, and physician assistants) (20) are often the first line of care for people living with chronic pain (20) and patients often do not receive chronic pain EBPs (14–16). In part, this is due to complex referral systems and insurance coverage (21) and a lack of pain specialists (4, 22). Therefore, understanding and integrating PCP perceptions of barriers to delivering these evidence-based therapies is a critically important next step toward improving implementation. Implementation science (IS) seeks to understand and close the gap between EBPs and clinical care (23) and may help to ensure that people living with chronic pain receive evidence-base management. The primary objectives for this study were to investigate PCP perceptions of barriers to evidence-based chronic pain management implementation, and PCP strategies (or facilitators) for addressing these barriers. The secondary objective was to develop a novel approach to using implementation science to address the evidence-practice gap in chronic pain management in primary care settings.

We selected a qualitative approach to answer our research question. Qualitative data collection and analysis support both positivist and constructionist epistemological perspectives (24). We used qualitative data collection (i.e., interviews) and analyzed data using exploratory or content-driven analysis which answers the question, “what do x people think about y?” (24). We conducted one-time semi-structured telephone interviews with PCPs between May 2019 and August 2019. The COREQ (COnsolidated criteria for REporting Qualitative research) Checklist was used and can be found in the Supplementary File S2.

We interviewed 21 PCPs (see Table 2). The mean age was 51 (SD = 10.8). Half of participants (n = 11) identified as male, and most participants (n = 17) identified as white, with three participants identifying as Asian. The average number of years post-training was 21 (range 3–35); participants spent an average of 3.3 days (range 0.5–5) in the clinic per week. Participants were mostly physicians with 61% (n = 13) being Medical Doctors (MDs) and 28% (n = 6) as Doctors of Osteopathy (DOs). The remaining two participants were advanced practice providers (one Physician Assistant and one Nurse Practitioner). Half of the participants worked in an academic clinic (n = 10), and most (n = 18) were involved in some form of teaching (e.g., supervising medical residents). Interviews were an average of 21 min 14 s (range: 15:58–25:05). No participants refused participation nor dropped out for the duration of the interviews.

Results from Steps 1-3 are as follows: Step 1. Identify determinants of clinician's implementation of chronic pain EBPs. Participants reported barriers to implementing chronic pain EBPs in primary care settings across patient, provider, and system levels. A major issue cited was access to non-pharmacologic treatment such as cognitive behavioral therapy and physical therapy. One participant described these barriers in this way:

Step 2. Identify strategies clinicians used in providing chronic pain treatment to overcome barriers. PCPs reported strategies they use to overcome these barriers such as using existing resources, rapport with patients, and understanding of the health system. For the barrier of access, some participants described developing in-house psychological treatment for people living with chronic pain. A participant described this approach used by their clinic to address access to psychological treatment for chronic pain management:

Step 3. Align Steps 1 & 2 with existing implementation frameworks. The PI (LEA) reviewed the barriers and strategies identified from our inductive coding of interview transcripts and mapped them onto existing CFIR domains and constructs and ERIC implementation strategies. These selections were then reviewed by co-investigators (LEA, MEH, SME, and JSM) for acceptability from a qualitative (MEH) and chronic pain (JSM) perspective. We did this by reviewing the definitions of CFIR domains and constructs and implementation strategies with the identified themes from the data and matched definitions. For example, the barrier to access aligned with the CFIR Outer Setting domain and construct of Patient Needs and Resources because the PCP recognized the access need of patients living with chronic pain. The strategy to overcome aligned with the ERIC implementation strategy of Change Service Site because they moved treatment to a setting easily accessible by patients (i.e., in-house treatment). In the interviews, participants did not suggest any strategies to address the patient-level barrier of co-morbidities nor practitioner-level barriers. Table 1 has a list of themes that arose through our data collection and analysis and their alignment with existing implementation frameworks.

Our results, while not generalizable, align with previous research within chronic pain management in primary care settings: the evidence-practice gap persists, particularly for implementation of non-pharmacologic approaches, despite ongoing efforts to integrate IS principles into chronic pain management (12, 28–34) and providers using creative approaches to ensure their patients receive the care they need. The importance of the therapeutic relationship between the patient and PCPs is well-known. The ongoing patient-practitioner relationship often helps to frame and interpret difficult conversations, including those about chronic pain (35). Conversations about chronic pain management can be difficult as PCPs may struggle to believe patients' reported pain (36) and patients may struggle to feel trusted in their own experience (35).

However challenging, participants in our study expressed a degree of hope and at times desire to push through the hard conversations to address a patient's pain or possible substance dependency. Chronic pain management still exists within the shadow of the opioid epidemic. The consequences of opioid use (and misuse) continue to be pervasive in national dialogue, continuing education, and funding—especially given the uptick in opioid-related deaths during the COVID-19 pandemic (37, 38). Again, given these challenges, across the board, practitioners in our study expressed the desire to navigate the complex challenges of treating chronic pain with their patients despite these contextual barriers. Participants demonstrated an immense desire to work with patients to navigate complex systems and manage their chronic pain. Unexpectedly, PCPs in this study were open to identifying and using non-opiate treatments to help manage their patient's chronic pain, even as these treatment options were typically harder to access. Participants showed surprising resourcefulness in overcoming known barriers to get patients the help they need, such as hosting chronic pain support groups in the primary care clinic. These strategies create an innovative foundation for future testing and implementation in more diverse clinical settings.

We developed a three-step approach as one way to interpret limited samples of provider experiences specifically within chronic pain to begin to address implementation gaps within a given clinical context. Our approach attempts to leverage the field of implementation science by providing a way for researchers and practitioners alike to address this gap by taking a strengths-based approach to the experience of PCPs and how they continue to find innovative ways to navigate complex chronic pain management. We then used the field of implementation science to complement and build on the front-line knowledge of barriers to treatment and the navigation of those barriers. Alignment of current barriers and strategies with the larger field of implementation science may provide avenues for additional strategy discovery or approaches used in other settings which can promote the ongoing evolution of navigating the complex barriers to treating chronic pain.

Clinicians, scholars, and quality improvement evaluators can leverage existing knowledge by aligning existing barriers and strategies with the larger literature. Future practice and research should focus on building on the existing model to include Step 4 tailor existing implementation strategies to the specific needs of chronic pain in the clinical context, perhaps by formalizing changes in service location (per our earlier example) and expanding it to include virtual support groups and teletherapy (see Figure 2). Based on our findings, approaches should be patient-centered and solicit guidance from people living with chronic pain and their families and support systems. Strategies to overcome barriers and the interventions themselves should be adaptable and tailored to the given clinical context, type of chronic pain, and patient preferences. Important considerations include the degree to which the policy setting may impact PCPs ability to navigate barriers to evidence-based chronic pain management. Some contextual factors may include the degree to which mental health services reimbursed at the same or similar rates to physical health services (mental health parity) or the impact of the expansion of Medicaid influence the proportion of people with insurance coverage.