Academic Primary Care Clinic Adopts New Paradigm for First-Line Treatment of Low Back Pain

Background

Low back pain (LBP) is highly prevalent, a leading cause of disability, ¹ one of the costliest of all medical conditions, ² and among the most common reasons for a physician visit by U.S. adults. ^3,4 Spine care in the United States has been plagued with a pattern of deteriorating outcomes despite accelerating costs, ⁵ and because LBP is the most common condition for which opioid analgesics are prescribed, ⁶ the crisis of opioid prescribing and misuse reflects in part a failure in the clinical management of LBP. Primary care physicians account for 45% of all U.S. opioid prescriptions ⁷ with back pain being a leading reason, as they see the majority of patients presenting with back pain. ⁸ However, current evidence-based clinical guidelines from the American College of Physicians for management of LBP list opioids as the last treatment option for back pain, only recommended when other treatments present substantial risk to the patient. ⁹ Current guidelines call for nonpharmacologic therapies such as spinal manipulation, acupuncture, and massage therapy as first-line treatments to be offered before prescribing medications. ² Conventionally taught primary care physicians are not routinely trained in nonpharmacologic therapies and are, therefore, largely unprepared to provide guideline-concordant care for LBP. The authors describe in this study the implementation of an innovative primary care-based model intended to address this deficiency.

Innovative Model of Care for Patients with LBP

In the Primary Spine Care (PSC) model, a PSC clinician, who is specifically trained to provide comprehensive first contact and continuity back pain management, is integrated on-site within the primary care environment, provides LBP care in collaboration with the patient's primary care clinician. The PSC skill set includes the ability to work in a primary care setting to deliver evidence-based clinical management, appropriately educate and motivate patients, and effectively prevent and manage disability related to back pain. A PSC clinician has training in at least one nonpharmacologic approach and familiarity with other first-line therapies. They are able to identify the majority of patients who may be appropriately managed without special tests or specialty referrals, and triages the minority of patients who require additional evaluation (e.g., radiographs, magnetic resonance imaging, or electrodiagnostic testing), more intensive management (e.g., multidisciplinary rehabilitation), or invasive procedures (e.g., injection or surgery). A PSC clinician always remains at the front line of care, so patients have a consistent “go-to” professional. The PSC clinician can monitor and facilitate progress and help patients navigate third-party payment systems that can be particularly complex for those with personal or work-related injuries. Practice according to the PSC model has been reported to decrease variability in care, decrease costs, and improve outcomes for patients with spinal pain. ¹ A recent study found that use of a conservative spine care pathway within a primary care practice was associated with reduced health care expenditures over a 4-year period with the cost of spine care in the intervention group decreasing at nearly four times the rate as controls (28.3% vs. 7.2% reduction; p < 0.001). ¹⁰

Embedding a chiropractor as a PSC clinician in a primary care team is substantially different from simply including a chiropractor in an integrative medicine clinic. In the authors' experience, in many if not most integrative medicine clinics where chiropractic care is offered, it is one of several options available to patients; the chiropractor typically provides parallel services and is not required to coordinate patient care with other clinicians in the facility. By contrast, the PSC clinician proactively coordinates care from start to finish with the patient's primary care physician. This approach is not only a revolutionary concept for conventional primary medical care—it can also serve as a model for more effective “integration” of integrative medicine clinics that offer primary medical care.

Implementation of the PSC Model

The Dartmouth-Hitchcock Primary Care Clinic is the flagship primary care facility of the medical center and is staffed by faculty of the Geisel School of Medicine at Dartmouth. With a medical staff of 15 family physicians, 4 general internists, 3 physician assistants, 5 nurse practitioners, the clinic had 45,000 visits in 2019. Before implementing a PSC model in the lead author's clinic in 2017, management of LBP was not based on the clinical guidelines.

To implement the PSC model in their clinic, as PSC clinician, the authors hired a Doctor of Chiropractic (DC) with training in primary care and hospital settings, acquired through an integrative clinical practice chiropractic residency and fellowship. The DC had received postdoctoral training in delivery of evidence-based PSC in a large urban Veterans Administration hospital, collaborating with primary care patient care teams, relevant specialists, and associated health care clinicians. Time was also spent training in the hospital-based spine center and in operating room observation. Upon completion of the hospital-based residency, the DC completed a fellowship in PSC in a multidisciplinary university-based integrative health clinic.

Model implementation was led by the Department of Community and Family Medicine. The rationale and explanation of the model were met with unanimous approval by the primary care clinicians and support staff. Two patient care rooms in the clinic were dedicated to and equipped for spine therapy. Patients presenting with LBP were seen by a primary care clinician and received standard care for LBP with most patients additionally being referred by the primary care clinician to the PSC clinician. Embedding the PSC clinician in the primary care team facilitated real-time two-way communication in managing LBP. He could easily share his findings, diagnosis, and plan with the referring clinician, in addition to internal communication through the electronic medical record. If imaging was indicated, the primary care clinician provided the order based on the PSC clinician's recommendation. The primary care clinicians felt this added care option provided an alternative to potentially prescribing opioids. As the comfort level grew with the new model, primary care clinicians increasingly permitted office staff to directly schedule patients with LBP with the PSC clinician as a first line of contact. The PSC clinician also coordinated care when indicated with physical therapists located on-site and with a primary care physician team member who performed acupuncture.

Clinical care provided by the PSC clinician differed from typical chiropractic treatment. Less imaging was ordered, less spinal manipulation was performed, and no electrical stimulation, traction, or ultrasound were used. Instead, there was more of an emphasis on using diagnosis-based decisions, including risk stratification tools that favored specific exercises and other approaches, such as manual therapy, motivational techniques, and patient education within the context of the biopsychosocial model.

Barriors to Model Implementation

Barriers to implementing the PSC model included (1) Explicit bias: the belief that providers other than medical physicians are ill-suited or untrained to assume such a role in primary care; (2) Structural bias: limited insurance reimbursement for nonmedical providers; and (3) Implicit bias: physician and administrator lack of familiarity with the PSC clinician's training, expertise, and competencies. Explicit bias was successfully addressed by communicating the evidence supporting the suitability of nonphysicians in treating pain from spine disorders. Structural bias was overcome because availability of the PSC clinician effectively reduced “leakage” of patients to external providers, and the institution realized cost savings in the care of the self-insured employee population. Implicit bias was pre-emptively tackled through one-on-one education of physicians and administrators using evidence-based literature. A fourth potential barrier, which the authors did not encounter, is the problem of perverse incentives that can plague fee-for-service payment models: resistance to adopting a lower cost clinical pathway due to the resultant loss of revenue from specialist care, imaging, and procedures. In the authors' system, the PSC model was welcomed because it had the potential to keep care “in-house” as well as reduce inappropriate referrals to the spine center, thus allowing spine specialists to focus on patients that required their specialized skills.

Impact of PSC Model Implementation

Implementation of the PSC model ^8,11,12 allowed the lead author's practice to make significant strides toward meeting the evidence-based standard of care for management of LBP. ^9,13 The model's impact was monitored as part of a continuous quality improvement effort. After 1 year, care delivered by the PSC model was associated with a trend toward reduced average annual per-patient expenditures for spinal imaging ($532–$653 less), spinal injections ($646–$973 less), and spinal surgery ($6,914 less) as compared with management through usual primary care. ¹⁴ Unfortunately, the data were insufficient to determine any impact on the prescribing of opioids. Outcomes measurement at 3.5 years after implementation is currently underway.

In an internal performance improvement survey, 88% of the primary care clinicians in the lead author's clinic reported referring their patients with spine-related pain to the PSC program, and 59% favorably changed their own imaging ordering habits due to the presence of the program. Eighty-one percent reported that the new model it made it easier for them to care for patients with spine pain, and 0% said it made it harder. Common comments disclosed a uniform satisfaction with the model, a high level of confidence in the PSC clinician's expertise, and 100% of primary care physicians accepted the PSC clinician as an initial portal-of-entry provider for patients with spine pain. Given high unmet patient demand, there is opportunity for a second PSC practitioner. The lead author's hospital spine center is supportive of expansion as more appropriate treatment delivered in primary care translates to fewer cases reaching the spine center that do not need interventional treatment.

Conclusion

New models of evidence-based spine care in primary care settings are urgently needed to meet clinical guidelines for management of patients with LBP. For most primary care physicians, it is a challenge to operationalize the new clinical practice guidelines for care of LBP. The authors have described their experience with the PSC model as one approach to meet this mandate and thus improve the quality of spine care for their patients. Replication of their experience does not necessitate hiring a chiropractor; for example, osteopathic physicians skilled in spinal manipulation or physical therapists with experience in a primary care setting and knowledge about first-line spine care therapies also could serve in this role. In the authors' experience, the addition of a PSC clinician to their primary care practice allowed them to provide evidence-based LBP care and reduce the cost of care while simultaneously achieving high physician acceptance in an academic medical environment that is well known as a center for research on conservative spine care. The authors encourage their colleagues at other academic medical centers to disseminate, implement, and evaluate new models of spine care in primary care settings.