Application of Algorithm-Based Treatment Approach to Low Back Pain in the Emergency Department

Abstract

Low back pain accounts for nearly 4 million annual emergency department (ED) visits, and patient outcomes following an ED visit for low back pain are poor. Additionally, only a small portion of patients visiting the ED for low back pain follow up with outpatient physical therapy within 3 months, despite established benefits of early versus delayed physical therapy referral such as improved patient outcomes, less opioid use, and reduced downstream health care utilization. Integrating a physical therapist directly into the ED care team can facilitate evidence-based guideline concordant care and improve patient outcomes, however, physical therapists who are staffed into this role from other settings may lack experience with evaluating and managing patients with acute low back pain. Additionally, there are several unique considerations of the ED care environment which may make existing treatment-based classification approaches difficult to apply in this setting, including physical constraints (eg, delivering care in stretchers and hallways), higher symptom severity and psychosocial stressors necessitating an emergency visit, and greater likelihood of alternative medical diagnoses (eg, kidney stone, aortic aneurysm) contributing to symptoms of low back pain. This perspective presents a modified ED treatment-based classification system (ED-TBC) for low back pain with 3 illustrative case examples. The ED-TBC for low back pain can be used to facilitate guideline concordant care, increase physical therapist confidence in evaluating low back pain in the ED, and reduce clinical practice variation.

INTRODUCTION

Physical therapists are increasingly positioned in emergency department (ED) settings to evaluate and treat a range of clinical conditions, such as low back pain.^1–3 While the emergency care setting is a non-traditional clinical context for physical therapists to practice, it affords a unique opportunity to intervene early in the symptom trajectory of low back pain and implement proactive treatment strategies to optimize patient outcomes. Possible benefits of integrating physical therapists into the ED clinical environment include improvements in pain-related functioning, reduced medication use, expedited outpatient physical therapy referral, and reduced downstream health care utilization.^4–10 These benefits are particularly salient given recent research demonstrating that both patient outcomes and health care utilization substantially differ based on the type of initial contact provider seen for an episode of low back pain,⁹ and that only a small proportion of patients presenting to the ED for low back pain follow-up with an outpatient physical therapist.¹¹

However, traditional treatment-based classification (TBC) approaches for low back pain may be difficult to implement in the emergency care setting. The ED is a unique physical environment in terms of clinical space (eg, care delivered on stretchers or in hallways), lack of specialized equipment (eg, adjustable plinths), and high prevalence of psychosocial stressors (eg, the initial point of care for trauma). Compared to other treatment settings, patients who present to the ED for low back pain also tend to have a higher comorbidity burden⁹ and symptom severity,^4,12 coupled with less resources (eg, insurance, transportation, housing) and lower access to primary care.¹³ Additionally, patients presenting to the ED often have a relatively short symptom duration and may not be able to accurately describe their low back pain symptomatology or present reliable exam findings.¹⁴ In 1 study of ED patients with acute low back pain, 2/3 of participants presented to the ED within the first 3 days of symptom onset—with 1/3 presenting on the same day of symptom onset.⁴

Given the increasing attention to ED-based physical therapist care models, there is a growing need to introduce an evidence-based treatment approach to low back pain in the emergency care environment—particularly to assist physical therapists who are newly placed in this environment from other hospital acute care settings. This may reduce unwarranted practice variation among physical therapists who are new to this environment and for those who may lack experience in the evaluation and management of patients with acute low back pain.

In this article, we describe a modified TBC system for the evaluation of patients with low back pain in the ED (ED-TBC) setting. This algorithm is employed in an ongoing randomized clinical trial¹⁵ and provide 3 accompanying case illustrations. The ED-TBC algorithm incorporates existing evidence on physical therapist interventions for diagnosing and treating low back pain as well as our pragmatic experience utilizing a dedicated “ED physical therapist” role in an urban academic ED over the last 8 years.

CLINICAL NEED FOR A MODIFIED TREATMENT-BASED CLASSIFICATION FOR EMERGENCY DEPARTMENT PHYSICAL THERAPISTS

Low back pain is a substantially heterogeneous clinical condition, and there are numerous interventions utilized by physical therapists to treat patients with low back pain.⁹ Thus, a large focus of inquiry over the last 3 decades has focused on identifying subgroups of patients based on signs and symptoms that might benefit from distinct treatment approaches. The original TBC described by Delitto et al in 1995 had 3 levels.¹⁶ In the first level, the provider triaged patients into medical management for serious pathology, a referral for psychosocial consultation, or independent physical therapy management. Then, only patients appropriate for independent physical therapy were further classified into 1 of 3 stages based on pain intensity and disability status. Patients in stage I with high pain severity and disability were evaluated and categorized into syndromes named after the intervention (flexion/extension/lateral shift syndrome, traction syndrome, immobilization syndrome, mobilization syndrome).

The TBC system was updated in 2007 by Fritz et al¹⁷ to incorporate additional evidence since the original version. After screening patients for any medical red flags in level 1, the system had 3 classifications: manipulation, specific exercise, and stabilization.^18–20 This update specified criteria for identifying patients who would benefit from stabilization or directional preference exercises, recommended removal of the traction classification and immobilization terminology, and used a best fit when no TBC could be selected based on the required criteria. These updates were relevant to patients with high pain severity and disability status characterized in Stage I of the original TBC.

In 2015 Alrwaily et al updated the TBC,²¹ enhancing the decision making for the 2 levels of triage, 1 at the level of the first contact health care provider and the other at the level of the rehabilitation provider. Decision criteria were established for the first level to evaluate and refer patients with red flags, medical co-morbidities, and central sensitization.²² The remaining classification options at the first contact health care provider level, self-management or physical therapist management, integrated risk stratification tools to identify patients who might benefit from psychologically informed rehabilitation techniques. This update also renamed the TBC level 2 classification as: symptom modulation, movement control, and functional optimization. Classification at level 2 is based on the patient’s pain severity, irritability, and disability status. Revision of the TBC in 2015 was extensive, broadening frameworks for interventions beyond symptom modulation and linking evidence from the Clinical Practice Guidelines for low back pain from the American Physical Therapy Association and Academy of Orthopaedic Physical Therapy.²³

Furthering the most recent TBC update by Alrwaily et al,²¹ we developed a modified TBC system specific to the emergency care environment (Figure 1). This was intended to standardize an intervention protocol for a randomized clinical trial (NCT04921449) of an embedded ED physical therapist intervention for low back pain,¹⁵ that we now employ routinely in our ED physical therapist practice. In this embedded ED physical therapist care model, a physical therapist is forward-positioned onto the primary ED treatment team (traditionally defined as the physician, nurse, and technician) to evaluate and treat all patients with low back pain on a routine rather than discretionary consultative basis. In the ED physical therapist care model, the identification of patients appropriate for “medical management” is performed by the ED physician in parallel with the ED physical therapist’s assessment. We removed the initial level of triage by only the first-contact health care provider to include a physical therapist to determine the appropriate management approach. Once this triage for immediate red flags occurs, the ED physical therapist uses the ED-TBC as appropriate to select the rehabilitation approach based on the patient’s individual presentation.

An ED-specific modification to the 2015 TBC system was needed for several reasons. First, In our experience, nearly all ED patients with a provisional low back pain diagnosis tend to fall within Alrwaily et al’s revised symptom modulation category where in the ED setting pain intensity and disability status for conditions seen are almost universally high.^4,12 We use the initial guidance for intervention of patients matched to symptom modulation²¹ with some additional re-organization reflecting our pragmatic experience. Second, by virtue of presenting to the ED, patients tend to have a higher degree of psychosocial risk factors and lower self-efficacy.⁵ Consistent with the TBC, we utilize risk stratification tools, though we retain patients with low psychosocial risk to also subgroup-matched TBC by the ED physical therapist. Third, there have been advancements in understanding of pain and changes in terminology. It is Increasingly understood that many individuals have pain with more than 1 pain mechanism present. Rather than triaging patients with sensitization or nociplastic pain presentations to more specialized medical management, these patients are retained for matched intervention. Lastly, modification of the TBC was made because patients in the ED have a higher probability of serious pathology. Thus when patients do not meet criteria for a specific rehabilitation approach in the ED, they are classified into the low back pain of unspecified origin^24,25 prompting re-consideration of non-musculoskeletal pain sources and co-morbidities in consultation with the ED physician. This iterative process serves as a safety net and re-emphasizes Alrwaily et al’s guidance for the ED physical therapist to remain watchful of red flags and determine whether there are physical and psychological co-morbidities that might necessitate medical management or co-management.

EMERGENCY DEPARTMENT TREATMENT-BASED CLASSIFICATION

The ED-TBC algorithm includes the ED physical therapist in the triage of patients for red flags and non-musculoskeletal pathologies in parallel with ED physician for a provisional low back pain diagnosis at the first level of triage (Figure 1). Then, all patients are screened for high-risk psychosocial features using a modified STarT (Subgroups for Targeted Treatment) Back Screening Tool^26,27 Because the majority of ED patients with acute low back pain present within days of symptom onset, and many ED patients stated they were unable to answer the STarT questions due to a total symptom duration of less than a day, we modified the STarT Back Screening Tool question stem from “Thinking about the last 2 weeks” to “Thinking about the current episode of low back pain” and removed “in the last 2 weeks” from questions 1, 2, and 9. With a high risk subscore ≥4 (Figure 1, Step 1), the ED physical therapist is prompted to consider evidence-based psychologically informed physical therapy techniques in lieu of or as a supplement to any ED-TBC classification-specific interventions.

Notably, the ED-TBC algorithm assumes a funnel design (Figure 1) beginning with the most commonly encountered ED-TBC category from our experience in the ED and proceeding to less common TBC categories. The criteria for each category are modified from the TBC revision in 2007 by Fritz et al (Table 1). Each classification guides the initial physical therapist interventions that may be used to reduce the patient pain severity and irritability, encourages self-efficacy, and addresses fear of movement by encouraging the patient to remain active (active rest). As appropriate, exercise is prescribed by the ED physical therapist based on the classification. For example, if a patient has low back and leg pain that is reduced or centralizes with repeated or sustained movements in extension or flexion, they are then instructed in extension- (eg, prone press up) or flexion-based (eg, single knee to chest) exercises in the ED and a home exercise program. If a patient cannot be placed into a specific exercise classification, they are then evaluated for the manual therapy: low grade non-thrust mobilization classification criteria; if the patient responds positively, they are instructed in the performance of self-management strategies including midrange range of motion exercises, positioning, and corrective body mechanics/motor control with activities of daily living (eg, getting in/out of bed, chair). Examples of patient response and interventions provided in each category of the ED-TBC are shown in Table 2. If a patient does not respond, they are then sequentially assessed with the criteria for stabilization, followed by manual therapy: high grade thrust/non-thrust mobilization/manipulation and then pain predominant interventions. The sequence of the algorithm is shown in the Figure 1.

The ED-TBC retains patients who are identified by the 2015 TBC as having central sensitization and consider this to be the penultimate pain-predominant intervention classification.^28,29 Following International Association for the Study of Pain recommendations, we use criteria for possible nociplastic pain, defined as “pain that arises from altered nociception despite no clear evidence of actual or threatened tissue damage causing the activation of peripheral nociceptors or evidence for disease or lesion of the somatosensory system causing the pain.”^24,25 The clinical criteria can be efficiently gathered in the ED setting (Table 1). Characteristically the pain duration for chronic low back pain is defined as greater than 3 months. We use the National Institute of Health Task Force definition of Chronic Low Back which is defined as a back pain problem that has persisted at least 3 months and has resulted in pain on at least half the days in the past 6 months.³⁰ A pain predominant intervention classification supports the use of evidence-based cognitive and behavioral physical therapy techniques (eg, active listening, patient-led goal setting), sleep and nutrition education, and recommendations for a referral to a multidisciplinary pain clinic.

Of note, the ED-TBC matched intervention categories are not mutually exclusive as some patients may present with criteria that meet more than 1 category; however, matched interventions are provided when the criteria are met and the patient has a positive response upon re-examination. If the patient is not responsive, the ED physical therapist may proceed in the logical examination progression to categorize the patient into a subsequently ordered matched intervention.

As a safety check, low back pain of unknown origin is used for patients that do not meet criteria of any of the previous categories and necessitates discussion with the ED physician for a non-musculoskeletal source to the patient’s clinical presentation (eg, kidney stone, shingles) or the possibility of a concurrent musculoskeletal and non-musculoskeletal diagnoses (ie, lumbar radiculopathy and a kidney stone). Once non-musculoskeletal origins of symptoms have again been ruled out, the physical therapist can categorize the patient into an ED-TBC by best fit, akin to the 2007 TBC. This step necessitates collaborative clinical decision-making and reasonable modification of the criteria to decide on a best fit ED-TBC intervention. If patients cannot be classified by best fit, then by default the interventions provided are those that apply to all patients (Figure 1, Step 3) in the final step of the ED-TBC.

Essential to the ED-TBC, there are critical interventions that all patients matched to an ED-TBC category receive (Figure 1, Step 3). Of high importance is the facilitation of a rapid referral to outpatient physical therapist management. Early physical therapist management of patients with low back pain has numerous benefits.^7,9,10,31,32 An expedited referral for physical therapy can be facilitated in several ways including: (1) use of ED administrative staff to set up an outpatient appointment; (2) use of a navigator who contacts patients after the ED visit to facilitate the appropriate referral; or (3) providing the patient with a prescription for outpatient physical therapy. The ED physical therapist may also re-emphasize the benefits of outpatient physical therapy for patients who have a physical therapy referral from a previous provider for the ED low back pain complaint that had not been previously fulfilled. Optimal strategies to overcome barriers to patients seeking early physical therapist management for low back pain after obtaining a referral are continually evolving. Also essential to the ED-TBC is that all patients receive education and instructions in strategies to remain active (Figure 1, Step 3). Education includes reassurance about the source of back pain, overall favorable prognosis, timeline for recovery, and coaching to reduce fear avoidant behavior and promote function.

CASE ILLUSTRATIONS

We provide the following case illustrations of patients evaluated by an ED physical therapist using the ED-TBC algorithm, with patient specific examination results summarized in Table 3. In each of these cases, the ED physical therapist evaluated the patient in parallel with the ED physician as part of an embedded ED physical therapist intervention protocol deployed within a randomized clinical trial of patients in the ED presenting with acute low back pain.¹⁵

Case A

History

A 60-year-old male with history of diabetes and hypertension presented to the ED with progressively worsening right hip and right lower extremity pain limiting full weight bearing, ambulation, and sleep. He reported that the pain began 2 days ago with mild intermittent pain in the bottom of his foot and had progressed to a constant severe pain in his right posterior leg and calf (Table 3). The patient reported that this leg pain felt as if it was simultaneously radiating to and from his right hip. He reported that the pain was initially aggravated with right lower extremity weightbearing and ambulation. It was reduced when lying supine with his right hip and knee flexed. He denied red flag signs/symptoms.

ED Physician Exam

The ED physician independently evaluated the patient and noted a positive straight leg test of the right lower extremity, and low back pain reproduced at end range passive range of motion of the right hip (flexion, extension, abduction, adduction). The ED physician identified right calf swelling corresponding with a Well score of 1; a serum d-dimer level was obtained to rule out a deep venous thrombosis and was subsequently normal.

ED Physical Therapist Exam Using the ED-TBC Algorithm

In parallel, triage of the patient for red flags was conducted by the ED physical therapist and a preliminary diagnosis of low back pain with leg pain was hypothesized. The ED physical therapist examined the patient using the ED-TBC algorithm. The salient examination findings are shown in Table 3. In summary, the patient presented with increased pain in the right lower leg with lumbar flexion limited to 75% of normal. Pain in the lower leg worsened as a result. With extension, the patient reported pain in the right lower leg was diminished. Repeated extension significantly reduced the right lower leg symptoms. There was a positive straight leg test on the involved side at 35 degrees of hip flexion that reproduced the patient’s concordant right lower extremity symptoms; a sitting slump test was also positive on the right. Neurological screen of lower extremity myotome strength, sensation and deep tendon reflexes ruled out probability of progressive neurological deficits and cord involvement (Table 3). Although diabetic peripheral neuropathy was suspected based on the stocking glove distribution of diminished sensation on the right. Given these primary findings, the patient was placed into the specific exercise classification. Prone extension with partial range press ups were performed in the ED. A positive response was noted with reduced distal lower extremity symptoms after 10 repetitions. The patient was educated on centralization/peripheralization of pain and signs and symptoms of cauda equina syndrome as a reason to return to the ED. A home exercise program including prone extension exercise progression was prescribed. Education on the positive prognosis for recovery was provided given the patient response to extension exercises despite the high risk subscore on the ED modified STarT Back questionnaire. He was educated on remaining active with walking and modification of activities that included changing positions frequently and bed mobility strategies. These findings were discussed with the ED physician team and the patient was successfully discharged to home with a prescription for naproxen, cyclobenzaprine, a methylprednisolone dosepak, as well as a written referral to outpatient physical therapy at a local clinic for follow up.

Case B

History

A 68-year-old female with a history of hypertension, hyperlipidemia, knee osteoarthritis, and osteopenia presented to the ED with atraumatic central low back pain (see Table 3) that started 5 days prior and had been progressively worsening. She reported that the pain was worse in the morning, with any repeated movement in the lumbar spine, and with walking. She reported that the pain was reduced with rest in a sitting or lying position, although being in any position for a prolonged period of time was not tolerated due to pain. She had a 16-year history of intermittent episodes of low back pain and a previous diagnosis of degenerative spondylolisthesis at L5-S1. She reported worsening symptoms despite the use of naproxen and cold and hot packs on her back. She denied red flag signs/symptoms related to cauda equina, cord signs, or recent infection.

ED Physician Exam

The ED physician evaluated the patient and noted left lower back paraspinal tenderness to palpation, without tenderness at the sacroiliac joint or hip. There was no reproduction of symptoms with hip passive range of motion. Lower extremity myotomal strength, deep tendon reflexes, and sensation were normal. The ED physician administered a lidocaine patch and given the patient reported medical history ordered a lumbar spine x-ray that was notable for a grade 2 anterolisthesis of L5 on S1 with radiolucent zones suggesting bilateral pars defects, associated degenerative disc disease and endplate sclerosis, and sclerosis of the lumbar facet joints.

ED Physical Therapist Exam Using the ED-TBC Algorithm

The ED physical therapist examined the patient independent of the physician exam. Using the ED-TBC algorithm it was noted that low back pain worsened in midrange with return to neutral from both flexion and extension. Symptoms did not peripheralize with movement and a straight leg raise test was negative. Lower extremity neurological screen including deep tendon reflexes, myotomal strength, and sensation was normal. The patient was unable to tolerate prone positioning. A suspected step-off deformity at lumbosacral junction was noted in standing. In side lying, the patient’s concordant low back pain was reproduced with posterior to anterior accessory motion at L4, worse at L5, and both limited by painful end-feels. No allodynia was present. There was increase in pain with manual therapy with low grade non-thrust mobilization/manipulation at L5. Not meeting the specific exercise criteria and a lack of positive response to manual therapy: low grade non-thrust mobilization/manipulation for pain, the patient was evaluated for the stabilization category. The patient was unable to tolerate prone positioning, therefore the Prone Instability Test was not performed. Given presence of mid-range pain upon return from flexion and extension, abdominal muscle contraction with both ambulation and with accessory motion testing at L5 in side lying were attempted and reduced the patient’s symptoms compared to without the contraction. With the patient’s age over 40 years, she did not meet at least 3 criteria for stabilization category. The patient also did not meet criteria for the subsequent categories manual therapy: high grade non-thrust/thrust mobilization/manipulation or pain predominant intervention categories. The examination findings did not fit ED-TBC matched criteria which supported low back pain of unknown origin in the algorithm. The examination findings were discussed with the ED physician. After consultation with the ED physician, other non-musculoskeletal diagnoses were not believed to contribute to the patient’s presentation. Thus, the patient was placed into the ED-TBC stabilization classification by best fit, due to findings of midrange pain upon return from flexion and extension active range of motion, pain which improved with abdominal muscle contractions during ambulation and with accessory motion testing, and imaging results supporting spondylolisthesis at L5 identified as the most painful segment. The patient was prescribed home exercises that included motor control training to improve pain with movement including engaging abdominal muscle contractions with sit to stand, bed mobility, and neutral spine posture using rolling walker that resulted in ambulation with less pain. The ED physical therapist also facilitated a referral to outpatient physical therapy which was considered favorable by the patient due to the pain severity, reduced pain response to intervention in the ED, and acknowledgement of the need for better self-management strategies. The patient reported pain improvement during the ED visit and was discharged to home.

Case C

History

A 63-year-old male without past medical history presented to the ED with left lower back pain. He stated that he injured his back during intense exercise 12 days prior, with progressive worsening of pain and development of tingling and numbness in his left upper thigh. He saw his primary care physician who prescribed ibuprofen, acetaminophen, tizanidine, and gabapentin, and referred him to outpatient physical therapy. The patient reported that neither medications nor physical therapy have reduced pain severity, and that the outpatient physical therapist suspected hip involvement. The night before his ED visit, his persistent pain became intolerable, waking him up from sleep, and was newly associated with nausea—prompting him to come to the ED for evaluation.

ED Physician Exam

The ED physician exam noted left lower back and left flank tenderness to palpation, left lower quadrant abdominal tenderness to palpation, and normal lower extremity strength, sensation, and a negative straight leg raise test. The physician’s assessment and plan noted a broad differential, including potential concurrent diagnoses of musculoskeletal low back strain, urolithiasis, and diverticulitis.

ED Physical Therapist Exam Using the ED-TBC Algorithm

The ED physical therapist examined the patient with a potential concurrent low back pain source of symptoms during the triage. The ED physical therapist assessed the patient’s response to lumbar active range of motion for a potential directional preference response. The patient’s pain in the low back, flank, abdomen, and anterior thigh worsened with lumbar active range of motion. Repeated motion in flexion and extension did not result in centralization or peripheralization of symptoms. A straight leg test and neurodynamic test of femoral nerve were negative with pain reproduced with hip flexion but no change with ankle or cervical motion. Accessory joint motion of the lumbar spine did not reproduce symptoms. The patient did not have examination findings (Table 3) to meet the criteria for either manual therapy categories or the stabilization category. The patient did have concerning flank and abdominal pain, high pain severity, and pain reproduced in the abdomen with lumbar and hip range of motion. However, exam revealed no allodynia or pain that was present for greater than 3 months for criteria to be placed in the pain predominant intervention category. Thus, the patient was considered to have low back pain of unknown origin. The physical therapist assessment was discussed with the ED physician, particularly the inability to place the patient in a clear diagnostic-TBC category and the need to re-consider the possibility of a contributing non-musculoskeletal diagnosis based on the pain location and behavior. Based on ED physician independent exam and physical therapist’s examination findings, a urinalysis was ordered revealing positive red blood cell count. The ED physician subsequently ordered a computerized tomography scan of the abdomen and pelvis with oral and intravenous contrast showing an obstructing 4 mm stone in the distal left ureter with mild hydronephrosis and delayed left nephogram, with moderate to large volume of perirenal and periureteral fluid concerning for forniceal rupture. Urology was consulted and recommended outpatient follow-up; the patient was discharged with hydrocodone/acetaminophen, tamsulosin, and a referral to urology physician. He followed up with a urology physician 8 days later for definitive treatment of the kidney stone via left ureteroscopy, laser lithotripsy, and placement of ureteral stent.

DISCUSSION

We present a modified TBC system for the evaluation and treatment of low back pain the ED, the ED-TBC, which may differ from other physical therapist treatment settings due to a number of factors. Notably, the ED-TBC differs from Delitto and Fritz’s foundational work and Alrwaily et al’s update^16,21 due to the low prevalence of low symptom severity “self-management” presentations, the high prevalence of psychosocial stressors, and greater likelihood of alternative non-musculoskeletal diagnoses in the ED setting. The accompanying cases demonstrate the application of the ED-TBC to 3 common patient scenarios and illustrate how it can be used as a step-by-step guide for classifying and treating low back pain, starting with the most common diagnostic classification in our experience (directional preference), and logically funneling down to less common.

In addition to providing an algorithm-based approach to the evaluation and management of low back pain (Figure 1, Step 3), the ED-TBC facilitates early physical therapist management for patients with low back both with contact in the ED and facilitation of a rapid outpatient physical therapy referral (Figure 1, Step 3). Additionally, the ED-TBC emphasizes that all patients matched to the ED-TBC intervention for low back pain should receive education the overall favorable prognosis of low back pain, typical time course for recovery, and instructions for active rest (Figure 1, Step 3). Education for the pain predominant intervention category is unique (Figure 1). The ED-TBC also reinforces an iterative consultation with ED physician for non-MSK potential source when the patient does not clearly meet an initial ED-TBC criteria. This safety net step improves the differential diagnosis process. Lastly, the ED-TBC integrates screening for high-risk psychosocial features to further identify and assist in appropriate management (Figure 1, Step 1) but retains these patients for classification. These recommendations acknowledge early evidenced-based physical therapist management is beneficial for patients at higher risk of developing chronic symptoms, and that the experience of low back pain is a complex manifestation of biological, social, and psychological factors. Although prior research has established the necessary content and clinical efficacy of patient education and instructions for active rest,³³ the current body of research on the ideal components of psychologically informed interventions, cognitive and behavioral physical therapist techniques—and their efficacy—is still under development.^34–37 Thus, although we suggest active listening and patient-led goal setting as motivational techniques to employ in patients with high risk psychosocial features,^38,39 we acknowledge that others may prefer alternative techniques and content that may prove to be equally or more efficacious as the literature evolves.^40,41 Similarly, although we selected the Keele STarT Back instrument to screen for high-risk psychosocial features, we modified the question stem from “in the last two weeks” to “this current episode of low back pain,” as many patients in the patients felt they were unable to answer the questions based on their total symptom duration of less than 1 day. Other screening tools may provide better means to stratify patients for high-risk psychosocial features as evidence in this area evolves.

Of note, the ED-TBC attempts to identify a matched treatment that provides an immediate patient response whether that be increased motion, functional tolerance, self-management strategies or a reduction in symptoms. The patient’s favorable response in the ED may be impacted by numerous factors including expectations and contextual factors that are non-specific to the matched intervention provided.⁴² Most importantly the ED-TBC includes a mechanism to facilitate a referral to outpatient physical therapist services.

As ED-based physical therapist care models are increasing across the United States,¹ the ED-TBC can serve as a useful algorithm for physical therapists without prior training or experience in the emergency care environment. This stepwise approach may have several advantages, such as assisting in the selection of evidence-based therapeutic exercises to benefit the patient, increasing physical therapist confidence in evaluating low back pain in the ED, aiding a referral to the ED physician for the diagnosis of non-musculoskeletal origin of symptoms, and reducing unwanted clinical practice variation. This algorithm may therefore be useful to physical therapists who rotate or are flexed to the emergency care environment from other acute care or outpatient settings and who may have limited experience managing patients with acute low back pain. Although it may seem appealing to utilize only physical therapists with advanced training (ie, residency, fellowship-trained) in an ED-based physical therapist care model, this is not always possible. We also recognize the expertise needed to be an ED physical therapist seeing patients with wide range of clinical conditions, including dizziness/vertigo, balance disorders, and older adult discharge planning.^39,43,44 Physical therapists positioned in the ED will therefore require a broad skillset and knowledge base, further reinforcing the need for formalized care algorithms to standardize clinical practice across a range of conditions.

Physical therapists practicing in the ED setting encounter alternative non-musculoskeletal etiologies with a chief complaint of low back pain. Although physical therapists have extensive training in identifying low back pain “red flags” indicative of more sinister underlying pathologies (eg, cord compression, visceral referred pain), physical therapists with little outpatient experience may have relatively less confidence in identifying musculoskeletal pain patterns from other medical cause of low back pain, such as an obstructing kidney stone, ectopic pregnancy, or malignancy. Thus, we designed the ED-TBC to emphasize that the physical therapist should consider alternative medical diagnoses if the history and exam findings do not meet the criteria for any given TBC category, or consider the possibility of concurrent diagnoses. Importantly, the ED-TBC algorithm assumes co-management of each patient with an ED physician in an integrated care team, who would be evaluating the possibility of medical diagnoses in parallel with the physical therapist’s assessment.

The ED-TBC has included a step for unspecified low back pain which serves as an additional safety net for patients who do not meet the criteria of the ED-TBC matched intervention categories. This necessitates a discussion with the ED physician to re-consider a competing or co-existing non-musculoskeletal diagnoses. There are challenges to a medical differential diagnosis in the ED for patients whom a nociplastic pain presentation exists, which is in part the reason to retain these patients in the ED-TBC matched intervention algorithm. We include pain-predominant intervention category, for a lack of a better term, as a distinct matched intervention for a predominant nociplastic presentation. For patients who match this pain-predominant intervention classification, we use the interventions identified with high-risk psychosocial features based on the ED modified STarT Back subscore. We recommend shared decision-making between the ED physical therapist and ED physician leveraging the individual expertise of each clinician while taking into account patient needs and preferences. Also noted, the ED-TBC care algorithm as described may not be applicable to the practice of “extended scope physiotherapists” who operate as the sole clinicians in emergency care encounters, as is the case in settings potentially within and outside of the United States.

Finally, the ED-TBC was informed by our substantial experience utilizing a dedicated ED physical therapist care model over the last 8 years at a single urban academic ED. Although we have informally confirmed the ED-TBC approach with others who have similarly pioneered ED physical therapy programs at their own institutions, the ED-TBC algorithm has not been externally validated in other hospital settings with differing work cultures, operational flow, and patient populations.

In conclusion, this ED-TBC is a modification of the legacy and modified versions of Delitto and Alrwaily’s foundational TBC models^16,21 and is not intended to critique or displace this prior work but rather to adapt an ED-specific approach to practically guide less experienced clinicians who aim to start an ED-based physical therapy program. Following a review the foundational TBC models,^16,17,21 we present a modified version of the TBC for the ED that includes: (1) initial patient triage by the ED physician and physical therapist to screen for non-musculoskeletal conditions contributing to the patient presentation; (2) utility of the STarT Back score, modified for the ED to identify patients with high risk psychosocial features and proceed to categorize the patient with the algorithm; (3) logical progression of clinical criteria used in the examination to match an appropriate intervention based on our pragmatic experience in the ED; and (4) facilitating a rapid referral to outpatient physical therapy for patients with low back pain. We believe this perspective will generate dialog and future research related to ED physical therapists’ contributions to the management of patients with low back pain in efforts to improve patient outcomes.

CREDIT—CONTRIBUTOR ROLES

Kyle J. Strickland (Conceptualization [equal], Data curation [equal], Investigation [supporting], Methodology [lead], Project administration [supporting], Visualization [supporting], Writing—original draft [equal], Writing—review & editing [equal]), Howard S. Kim (Conceptualization [equal], Data curation [lead], Formal analysis [lead], Funding acquisition [lead], Investigation [lead], Methodology [equal], Project administration [lead], Resources [lead], Supervision [lead], Validation [supporting], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Amee L. Seitz (Conceptualization [equal], Data curation [supporting], Funding acquisition [supporting], Investigation [supporting], Methodology [equal], Resources [supporting], Validation [supporting], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]).

FUNDING

This work is funded by the US Agency for Healthcare Research and Quality R01HS027426 (H.S.K.) and the Doris Duke Charitable Foundation (COVID-19 Fund to Retain Clinical Scientists) (H.S.K.).

ETHICS APPROVAL

All physician and patient participants gave written informed consent to study participation for examples use in this perspective with the parent study approved by the Northwestern University Institutional Review Board (STU00213134).

ROLE OF THE FUNDING SOURCE

The funders played no role in the design, conduct, or reporting of this work.

DISCLOSURE AND PRESENTATION

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.

H. Kim receives a stipend from the American Medical Association as a Deputy Editor at JAMA Network Open. A. Seitz serves on Journal of Orthopaedic and Sports Physical Therapy Editorial Board, the Academy of Orthopaedic Physical Therapy Board of Directors, and receives support to attend meetings. A. Seitz receives salary support from the Eunice Kennedy Shriver National Institute of Child Health and Human Development NIH (R01HD098698) unrelated to this work.

This paper was presented in part as a platform presentation at the 2020 American Physical Therapy Combined Sections Meeting, February 12–15, 2020, Denver, Colorado, USA, and as an education session at the 2024 Combined Sections Meeting, February 15–17, 2024, Boston, Massachusetts, USA.

REFERENCES