ORIGINAL ARTICLE
Year : 2019 | Volume
: 2 | Issue : 2 | Page : 77--87
The impact of a complex consulting process with physiatry on emergency department management of back pain
Andrew John Haig1, Bradley Uren2, Sierra Loar1, Katrina Diaz2, Melissa Riba3, Kerby Shedden4, David Share5,
1 Haig Consulting, Londonderry, Vermont, USA
2 Department of Emergency Medicine, University of Michigan, Ann Arbor, Michigan, USA
3 The Center for Healthcare Research and Transformation, Ann Arbor, Michigan, USA
4 Department of Biostatistics, University of Michigan, Ann Arbor, Michigan, USA
5 Department of Medicine, Blue Cross Blue Shield of Michigan and the University of Michigan, Ann Arbor, Michigan, USA
Correspondence Address:
Andrew John Haig
Haig Consulting, P.O. Box 524, Londonderry, VT 05148
USA
Abstract
Objectives and Design: Prospective interventional trial. Setting: An university hospital emergency department. Patients: Two hundred consecutive persons admitted for back pain before, during, and 6 months after the process was begun. Interventions: Institution of a coordinated process of Emergency physician and patient education, standardized intake and order protocols, and rapid access to Physiatry and physical therapy. Patient characteristics, ED evaluation, ED orders, and 1-month rate of return were measured. Main Outcome Measures: Change in medication, imaging, referral to Physiatry and therapy, and 30-day readmission. Results: Most patients had red flags for danger and yellow flags for disability risk, 19-25% had objective radicular signs, and 14-24% had a second non-spinal complaint. There were no important demographic or pain characteristic changes. Compliance with study paperwork was low (20% patients, 6% physicians). There were significant increases in Physiatry and therapy referrals and a decrease in work restrictions. Medication prescriptions decreased overall, notably NSAID and diazepam prescriptions. Imaging orders did not change significantly. Return visits to the ED for back pain decreased from 16% to 4%, P = 0.02. Physician belief that patients had sciatica decreases (10% to 3%, P = 0.02), and detection of actual dangerous disease increased (3.5% to 6.5%). Conclusions: Even without full compliance with the protocol, this complex consultation process changed emergency management of back pain, most significantly by detecting more dangerous diseases, altering medication prescription habits, and decreasing return visits.
How to cite this article:
Haig AJ, Uren B, Loar S, Diaz K, Riba M, Shedden K, Share D. The impact of a complex consulting process with physiatry on emergency department management of back pain.J Int Soc Phys Rehabil Med 2019;2:77-87
|
How to cite this URL:
Haig AJ, Uren B, Loar S, Diaz K, Riba M, Shedden K, Share D. The impact of a complex consulting process with physiatry on emergency department management of back pain. J Int Soc Phys Rehabil Med [serial online] 2019 [cited 2023 Mar 31 ];2:77-87
Available from: https://www.jisprm.org/text.asp?2019/2/2/77/262680 |
Full Text
Introduction
Back pain is a huge drain on the American economy, estimated to cost as much as $90 billion direct costs per year and likely many times that much in indirect costs.[1] As an important portal of entry into the healthcare system, the emergency department (ED) is responsible for identifying and managing the many social and medical factors that impact the huge financial cost and personal suffering related to back pain.[2] Back pain comprises 3.15% of all ED visits.[3]
Taken logically, there are a number of mismatches between the needs of persons with back pain and emergency management. Most back pain presenting to an ED is not an emergency.[4] Urgent diagnostic test have a very high cost and a very low yield.[4],[5] Many of the more effective treatments for back pain, ranging from physical therapy to multidisciplinary rehabilitation to spinal injections are not readily available in the ED, and do not provide immediate resolution of symptoms.[6] Finally, an unpublished survey performed in preparation for the current study found that 22% of emergency medicine physicians had any exposure to physiatry during medical school and only 1 of 56 emergency medicine specialists had any physiatry exposure during their residency. Since emergency medicine clinicians are not required to receive any formal physiatric training they have little exposure to expert management of nonsurgical spinal disorders. Despite the substantial volume of back pain cases presenting to them, there may be more immediate rewarding challenges in their practice. In the end Emergency Medicine Physicians appear to have lower than optimal competency and the motivations to become proficient may be lacking.
The predictable consequences of these mismatches are evident. Emergency physicians have been shown to deviate substantially from good care; ordering too many tests and medications, while not effectively dealing with the psychosocial factors that greatly influence outcome from back pain.[4],[7],[8] The cost of care in the ED can be substantial. A study in Maine found that the cost of a visit for back pain ranged from $400 to almost $2000.[9] This figure does not include the costs of misdirected future care, lost work, and other expenses. Nevertheless, given societal perception of back pain as an emergency, the often acute severe presentation, the strong psychological correlates of chronic pain, and the real, though uncommon possibility of emergent, dangerous etiology, it is unlikely that back pain management will be shifted away from EDs in the near future. Emergency medicine must find a better way to manage back pain.
Changes in back pain management are not easy and not simple. A recent Cochrane report on interventions to improve appropriate imaging for back pain characterizes a number of potential interventions, including audit and feedback, distribution of educational materials, educational meetings, educational outreach visits, reminders, provider incentives, case management, and clinical multidisciplinary team formation.[10] This review concluded, “For low back pain studies, no intervention evaluated to date produced convincing effects for improving the use of lumbar imaging.” Since then research has shown that a process that removes decision-making from primary care and places it in the hands of PM and R physicians can change the rate of spinal imaging.[11] It seems reasonable that an approach that facilitates such intervention in the ED might have positive results.
There are numerous forces at play in the management of back pain in the ED. Among these are the intake and patient triage process, physician knowledge, skill, and attitude, patient education, and access to appropriate follow-up care. A plausible approach would view the problem in a holistic way, with coordinated alterations in each of these areas. We have designed a reproducible, yet complex, and multifaceted organizational consultation intervention for back pain management in the ED. The current study measures changes in back pain management before, during, and after implementation of that program in a university hospital.
Materials and Methods
This is a prospective study analyzing the impact of a complex back pain management process at a Level 1 Trauma center in a university hospital. The study was approved by the Institutional Review Board of the university.
[Figure 1] illustrates the study methodology. The project involved review of 200 records before the intervention, immediately after the launch of the intervention and at 6 months after launch. Records of persons aged 18–80 who were admitted to the ED with a diagnosis of low back pain, including associated terms such as sciatica, lumbar disk herniation, lumbar spinal stenosis, back strain, and sprain were included. This search included persons with discharge ICD9 codes 353.1, 353.4, 720.2, 721.3, 721.4, 722.10, 722.52, 722.73, 722.83, 722.93, 724.02, 724.2, 724.3, 724.4, 724.5, 724.6, 724.7, 724.8, 724.9, 847.2, 847.3, and 847.4.{Figure 1}
The intervention included codified programs for back education and special program duties for ED staff, ED physicians, physical therapists, and physiatrists. ED triage staff were instructed to give all patients with back complaints a clinical spine questionnaire and provide to the clinicians a check-off history, physical examination, and order sheet. Negotiations occurred to have Physiatrists and physical therapists agree to see specified patients within 48 h. ED, physiatry, and physical therapy office staff developed a triage process that ensured appropriate access. The same protocol was successfully launched at another tertiary care community hospital and a distant smaller community hospital. Data from these sites are not available for study. However, this process demonstrated reproducibility of the process in different settings.
Staff education
Staff education began with a planning meeting with the ED leadership. An ED champion (BU) was assigned. Subsequently, approximately 20 min was set aside at an ED physician meeting for the principle investigator to review the protocol and answer questions. On the starting date of the trial, and over the subsequent weeks physicians received in their mailbox a description of the program, a copy of the patient brochure, and 4 “case of the week” sometimes humorous case descriptions that presented best practices and modeled optimal patient communication. A video and accompanying description of the adequate back pain physical examination was designed and made available. ED nursing staff were provided with an in-service educational talk. The principal investigator and site leader visited the front desk staff occasionally to remind them of the project. Posters with the program logo and themes were distributed in the ED staff areas.
Patient education
The importance of patient education was emphasized to the clinicians. They were encouraged to tell patient that: (1) back pain is usually not dangerous. (2) It is important to keep active, and (3) diagnostic tests are usually not helpful. A patient pamphlet was developed to include information on diagnosis, treatment options, prognosis, and consumerism, along with four basic back pain exercises. Therapy interventions, as noted below, further emphasized this education.
Routine forms
During an individual patient encounter an intake form was to be given to any patient admitted to the ED with back pain complaints. The form served to ensure that physicians were aware of critical medical or psychosocial factors, to encourage them to perform an appropriate examination, and to make it easier for them to order appropriate tests than to take nonevidence-based or cost-ineffective routes. It begins with a two page patient questionnaire that includes demographic information, a pain drawing, and checklists for “red flags” (dangerous problems) and “yellow flags” (risk for disability). [Table 1] lists red and yellow flags used in data tabulation and physician handouts. All of these were not included in the brief patient questionnaire. The patent questionnaire is followed by a 1-page checklist physician examination form that covers pain reproduction, neurological deficits, and abnormal pain behavior. Following this is a checklist order form with four forced choices tied to specific diagnostic and treatment options. [Table 2] is that checklist. Although physicians were free to opt out of these treatments, they were encouraged not to do so without rationale. Because the hospital was preparing for transition to a new electronic medical record system, the order form could not be used as an official order set during the study, however physicians were asked to consider it an “order planning sheet.” The intake form was distributed by the front desk staff and collected by a research assistant at completion of each week. Data for the study were not gleaned from this form, but from the formal medical record generated by the physician.{Table 1}{Table 2}
Therapy protocol
For patients with straightforward, but significant acute back pain the program recommended that physicians refer to physical therapy. A 4-visit physical therapy protocol was taught to the therapists at the hospital. This intake assessment looked in more detail for risk of disability. The specific therapy treatment technique was left to the discretion of the therapist. Therapy ceased after a maximum of 4 visits, though patients did not have to attend four visits. At discharge the therapist reassessed risk for disability or dangerous disease. If the therapist felt the patient required more treatment they could opt for referral to the physiatry spine clinic, which committed to see these referrals within 2 business days.
The physical therapy component of the process required special external structure. Therapy referrals are not a natural thing for most emergency medicine physicians, who have little training in physical therapy methods and who generally do not want to have responsibility for care beyond their shift. Therefore, although the ED physicians were legally responsible for therapy orders, all therapy questions or problems were to be directed to the local physiatrist, who promised to see the patient in follow-up if needed. Therapists have the time and expertise to educate patients about back pain. However, licensure, limitation in professional knowledge, or fear of offending referring physicians often interfere with physical therapists fully educating patients about back pain. Therefore, the therapists were required to review a physician-authored patient pamphlet that addressed all essential aspects of back care including good prognosis, diagnostics, medications, injections and surgery, in the context of consumerism. Since isolated physical therapy is not sufficient for the management of chronic pain, any chronic patient erroneously sent to physical therapy for the protocol was offered a physiatry consultation and therapy ceased.
Physiatry consultation
ED physicians were encouraged to refer to physiatry patients with any problem that concerned them (e.g., concern for dangerous disease, concern for risk of disability, severe or complex pain, chronic pain, recurrent visits to the ED, etc.). The physiatrist spine program committed to see all protocol referrals within 48 h. This was operationally accomplished through faxed patient history forms with referral notes, scheduling in the university hospital's computer system, and set appointment “slots” that were often double-booked due to a high anticipated no-show rate.
The protocol operated within the usual billing policies of the physician and therapy practices. Uninsured persons were encouraged to see the physiatrist or therapist but were informed of costs. Some local insurers waived physical therapy pre-approval for persons referred through the process for the duration of the project.
Methods and measurements
University hospital medical records of the patients were reviewed and coded. The data recorder (KD) was an experienced coder employee of the ED. She underwent a 3-h training session with coders from other potential research sites. The coders examined 10 sample charts, and this coding was compared to the principal investigator's coding. Of 4650 items coded, there was a 3% disagreement, with most disagreements correctable by refined definition or clarification. A final data dictionary and “Frequently Asked Questions” document were made and shared with the coder. Many of the codings were derived from tables in the University of Michigan Practice Guidelines for low back pain.[12],[13] Most pertinent was the definition of “inexpensive” and “expensive” NSAID's.
To maintain independent data collection and analysis, the coder was an established employee of the ED with direct supervision from a research administrator who had no affiliation with the project. The database was sent by the hospital employee to the independent statistician (KS) who was paid by the study sponsor through university rather than by the consultant/investigator.
The approved study protocol allowed review of records for return to the ED in the month after their initial visit; however, it did not permit the investigators to evaluate the record to determine whether patients actually attended physical therapy or physiatry consultation.
Outcomes
Key outcomes included changes in physician diagnosis, imaging, medication prescriptions, and discharge follow-up recommendations. Return to the ED within 1 month.
Analysis
The data were cleaned and checked for errors. Statistical analysis was performed using an independently funded statistician. Comparisons between nonparametric statistics used the Chi-square, whereas parametric comparisons used ANOVA.
Results
It took 54, 30, and 31 days to obtain the 200 pre-, 200 during, and 200 post-intervention cases. Notably, the educational interventions and processes were launched simultaneously with the 1st month of data collection during which there was a faster than expected subject recruitment. Thus as [Table 3] shows, only 20% of the first 200 targeted patients after intervention filled out the intake forms. At 6 months (after active promotion had ceased for 3 months), only 6% of patient intake forms were turned in. The number of project forms filled out and filed by clinicians and was also small, but curiously increased at 6-month follow-up compared to during intervention [Table 4].{Table 3}{Table 4}
The populations treated at the three-time frames were similar in terms of demographics and medical presentation, except for some small changes in payer source [Table 3] and [Table 4]. However, after the intervention began the descriptions of the pain changed, with a significant decrease in the perception that pain was radiating and a borderline significant increased recognition of cases of chronic pain.
Physician diagnosis
The clinical interaction and findings are described in [Table 4]. Most patients were thought to have mechanical pain, and the number of persons thought to have sciatica decreased with the intervention. This is despite a nonsignificant increase in objective physical examination findings attributable to radiculopathy.
Systemic causes of pain were considered most likely in 7%–10% of admissions. Red flags for dangerous disorders were ubiquitous, occurring in about 2/3 of patients. Detection of red flags, in general, did not increase with the institution of the protocol. However of the 17 red flags listed, detection of four key red flags (diabetes, bedtime pain, steroid use, and recent bowel/bladder change) increased significantly (all at P < 0.005) with subsequent less detection at the 6 month period for all but the bowel/bladder flag (details not shown). Paralleling the pattern of increased key red flags, the number of confirmed dangerous disorders (cancer, spinal infection, and fracture) detected during the ED admission increased from 3.5% to 6.5% during the protocol, then dropped somewhat to 5.5% after active intervention ceased.
The detection of yellow flags (again ubiquitous at 73%–82% of patients) did not change during the study, nor did observation of excessive pain behavior or diagnosis of psychiatric diseases. Among the specific yellow flags, there was a statistically significant (all P < 0.005) decrease in physician detected psychiatric diagnoses, increase in detection of legal issues, and increase in detection of “more than two previous musculoskeletal complaints” (details not shown).
Clinicians actively diagnosed or treated a second complaint not related to the spine in 14%–24% of persons seen for back pain. These were primarily other musculoskeletal, neurological, psychiatric, and infectious disease complaints.
Imaging and laboratory tests
A drop in magnetic resonance imaging orders did not reach significance and was countered by an increase in computed tomography, also not statistically significant [Table 4]. Although the number of spine X-rays ordered did not decrease the number of images per study decreased significantly from around 5 to 4. This might be attributable to education that emphasized that 5-view film series have little to offer compared to three view lumbar spine X-rays. Blood or urine testing related to spinal disorders included blood counts, sedimentation rates, urinalysis for infection. These increased to more than half of cases but other tests did not increase significantly. “Nonspinal tests” including things such as electrocardiogram, ultrasound for pregnancy or X-rays of unrelated body parts, remained about the same at 9%–11% of cases.
Medication prescriptions
Medication use changed substantially through the study, with decreases in the prescription of all types of medication [Table 5]. Notably the use of “muscle relaxers” and diazepam decreased. The habit of prescribing more than a week's supply of an opiate seemed to drop during the study, returning to baseline at 6 months.{Table 5}
Discharge follow-up recommendations
Discharge planning and outcomes are shown in [Table 5]. Throughout the 600 cases in this study, no emergency physician requested specifically that a patient follow-up with themselves or an ED colleagues. Physical therapy consultation was never prescribed in the pre-intervention period but was ordered in 15%–19% of cases afterward, with the protocol name mentioned specifically. Physical Medicine and Rehabilitation referrals went from 7% to over 20%. There was a decrease in nonspecific “ED as needed” follow-up instructions. Actual attendance in therapy or physiatry clinic was not measured. A large increase in admissions to the hospital is likely an artifact of the coincidental launch of a new medical observation unit causing characterization of some patients as “admitted,” during the later parts of the study. There was no decrease of time spent in the ED.
Clinicians were encouraged to counsel patients to keep active and to avoid work restrictions. This appeared to have an effect on discharge advice, with a decrease in formal restrictions. The medical record showed a decrease in recorded activity advice, though actual patient interaction could not be studied.
Return to the emergency department
A probe into the medical record to see whether patients had returned to the ER within 1 month after admission showed a 75% decrease in “bounce back” patients from an original 16% eventually to 4%.
Discussion
This study revealed the effect of a multifaceted approach to improving the care of persons with back pain in the ED. The program appeared to impact both the workings of the ED and its output in favorable ways. The methodology and specific findings should be understood, and the implications for policy and practice should be discussed.
Specific findings
To our knowledge, no prior study has characterized the history, physical examination, diagnostics, and treatment plans for back pain in an ED to this extent. While prior studies have examined the impact of physical therapy on emergency department of back pain, this is the first to provide comprehensive support beyond the limited licensure of a therapist.[14]
One important finding is that most patients presenting to the ED for back pain have some “red flags” and “yellow flags,” which can be triggers for action. Many also have second reasons for seeking care. Back pain guidelines are typically designed by primary care physicians and spine experts, but not emergency physicians. The current data suggest that these guidelines may not reflect the complex reality of care in the ED.
The work of emergency physicians is made even more complex by their practice patterns. Faced with numerous red flags that could justify diagnostic tests, patients with real physical examination findings, patients with multiple complaints presenting simultaneously, an indigent population who may not have insurance to seek appropriate follow-up, and very few people whose findings actually require urgent management, the emergency physician who has only one crack at the case is under more pressure than others to use clinical judgment rather than evidence-based protocol alone. This choice of never following up on cases, as seen in the current population, is not written in stone. However, it seems unlikely that emergency physicians will seek longitudinal responsibility for patients. Hence, the pressures on ED physicians to make snap judgments rather than allowing a case to unfold over time will likely continue unless these complex cases are handed off to someone who is willing to follow them.
Physician education in this project may have improved diagnostic impressions.
The number of patients thought to have “sciatica” dropped during the intervention period. Since there was no change in the number of patients with objective physical examination findings for disk herniation, this date suggests the possibility that the pre-intervention physicians were misdiagnosing sciatica. Before this intervention, we performed a survey study of this physician group and the two community hospital emergency medical staff. (unpublished) Presented with an unambiguous case of acute nonradiating back pain, 7% of these emergency physicians labeled it as sciatica. In the differential diagnosis, over 50% felt the case could reasonably be attributed to disk herniation and 20% to spinal stenosis, neither of which was a plausible explanation.
Some of the physicians appear not to be familiar with the diseases that cause back pain. Previous work shows that emergency physicians are three-fold more likely to deviate from guidelines regarding sciatica compared to an index population of occupational medicine physicians.[15] Although acute management of sciatica should not deviate much from acute management of mechanical back pain, this diagnosis can lead less-educated ED physicians to order unnecessary tests, provide undue advice for caution and create physician-ordered disability.
Confirmed dangerous disease such as cancer or infection was found in only a small percentage of patients. However, it is worrisome that the percentage of confirmed dangerous diseases increased during the intervention. Were bad diseases missed before the protocol was in place? Despite the ubiquity of “red flags” in general, during the intervention physicians increased their detection of some key red flags: bowel and bladder change, night pain, diabetes, and steroid use. Perhaps the routine paperwork and an increased educational focus increased the detection of “real badness.”
The dangerous spinal disease can happen. However for many individuals, for society and the economy, the most devastating spinal disorder is “disability.” Physicians improve long-term outcome by counseling patients not to be afraid and keep moving and cause disability by scaring their patients.[16] Risk of disability is easily detected in most cases.[17] Advice to keep active and not be afraid; avoidance of unnecessary work restriction, imaging and surgery; and education in self-treatment can be important interventions to prevent disability. The study data show a decrease in activity restrictions. It is quite possible that the physician conversation with the patient changed, and the patient hand-out may have had some value. However, these could not be evaluated in the current study. Interaction with physical therapists trained to counsel in this area and with the Physiatrists may also have had an unmeasured positive impact.
Most guidelines recommend intervention based on time with back pain characterized as acute, subacute and chronic. However, it is apparent from [Table 4] that many or most “acute” back pain episodes have evidence of chronicity or recurrence. Another study found that 3 months after ED discharge, around half of back pain patients reported functional impairment, moderate-to-severe pain and analgesic use within the previous day.[18] Physicians who treat “acute on chronic” pain as if it were acute pain may be headed in the wrong direction.
Management of chronic pain in the ED is especially concerning since none of the evidence-based treatments for chronic pain are appropriately initiated by a physician who will not follow-up on the case. Evidence-based and highly successful management of chronic pain involves multidisciplinary teams and long-term follow-up. Other treatments that are occasionally appropriate include surgery, injections, or long-acting medications. Yet Jorgensen stated, “The ED's mandate is immediate symptom relief, not follow-up care and coordination of healthcare resources.”[9] The mismatch between this philosophy and the appropriate treatments for chronic pain raise serious questions. Should doctors who do not intend to engage in rehabilitation, perform surgery, spinal injection, or provide permanent restrictions ever order a diagnostic test for a disorder they suspect is not acutely dangerous? Should physicians who take no responsibility for long-term care ever initiate or continue the use of addicting pain medications? When patients present with “acute on chronic pain” (which is, in the final analysis, chronic pain) it may not feel good to send the patient home without treatment. However, that may be the right thing to do. Declining to initiate any treatment and instead offering referral to an expert is the most appropriate action for chronic pain that presents to the ED.
Overall there was a decrease in medication prescriptions. Prescription of diazepam and “muscle relaxants” (which have no direct effect on muscle) decreased. It was disappointing that certain parameters did not change. Opiates were still used frequently despite evidence that they are not more effective for acute back pain and that they are a cause of many overdose deaths and addiction problems.[19] MRIs, X-rays, and CTs were still ordered despite their limited utility in the ED. The duration of ED stay did not change significantly. It seems that whole-system redesign is needed to take advantage of the process, but even a decrease in imaging, which did not occur, might have substantially cut time in the ED. Interviews with physicians in the ED suggest that not all see protocol patients as worthy of breaking old habits. If crowded EDs are to improve throughput, more stringent application of the intake and discharge protocol along with increased physician education and peer review will be needed.
It is unfortunate that we could not capture the most important part of a back pain visit—the discussion with the patient. Advice to be careful or modify activity has been shown harmful while encouragement to have no fear, continue usual activity and expect recovery is the only intervention shown to improve long-term outcome from acute back pain.[16] The emergency physician who has not been trained in valid communication about back pain, such as that proven effective in Aage Indahl's landmark study, can do serious harm.[16] A physician's order to limit work can actually cause a person to become chronically disabled.
The most telling finding is the 75% drop in “bounce back” cases-those who returned to the ED within a month. Return visits are not desirable in terms of patient satisfaction, cost, or ED workflow. Although the study does not provide patient outcomes, this drop in return visits likely represents a substantial improvement in patient care. More appropriate diagnoses, fewer medications, and fewer work restrictions all suggest that quality was improved.
Methodology and comparisons
Most high-quality medical research involves focal and specific interventions, controlling as many variables as possible. However, the management of back pain in an ED is not a simple linear process. It is not logical to expect that any single intervention such as a drug, pamphlet, or therapist would have a great impact. A more realistic intervention would be complex. Research on complex processes as “black boxes” can be performed; however, the black box needs to be credibly reproducible if the findings are to be generalizable.
There are limitations to the current study. A university hospital presents a reasonable site for such research. However, the experience in one institution may not apply to other settings. One important challenge in this kind of research is to ensure that the complex intervention is practical and reproducible at other sites. The protocol was designed for reproducibility, including physician and patient educational modules, posters, forms, and processes. The program was also successfully launched at a large urban hospital and a smaller community hospital at the time of this research project. This demonstrates that the protocol is not just viable in an academic setting. Thus, while viability or reproducibility of the protocol may be demonstrated, studies of outcomes at different settings will be an important next step.
In general, a pre-post intervention study can show quite a bit, but cannot control for unexpected events. Two changes in this study include the introduction of a medical observation unit resulting in a drastic increase in hospital admissions, and the inability to hard wire the physician exam and order forms into the medical record because of a transition to a new electronic record system. The first made it impossible to determine whether the protocol could decrease hospital admissions. The second likely was an important factor in physician noncompliance with some recommended guidelines.
The small number of forms filled out during intervention is likely because the first 200 cases came so quickly after the protocol was launched. It is likely that compliance (and perhaps results) improved in the subsequent 3 months, but that could not be measured. Furthermore, during the transition to an electronic medical record, these forms could not be useful documents that would be entered into the record as patient history, physical examination and order sets.
The protocol required the principle investigator to completely step out of the process after the first 3 months. The data and our subjective observation have shown us that the intervention loses momentum if there is not an administrative process in place to ensure continued use of forms, physician and staff education, and smooth referral to the Physiatrists and therapists. One year after the principle investigator stepped completely out of the program a committee of emergency medicine, medical observation unit and physical medicine and rehabilitation clinicians and administrators met to propagate the process, which continues to this day.
It is unfortunate that the study was not able to follow patients in more detail. Actual patient satisfaction, percentage of patients who actually saw physiatry or physical therapy, and long-term cost measures such as imaging, surgery, and work disability are important variables that could not be assessed.
Despite any limitations, this study provides a unique view of the ED and an innovative strategy for the management of back pain there. We could find only one previous study that proposed an ED intervention for back pain. Lao found improved pain and patients' satisfaction in ED patients who were randomized to see physical therapists.[20] Lao's intervention addressed only one aspect of good spine care. There are limits to physical therapy intervention for back pain. Not generally within the scope of physical therapy practice are appropriate to use of medications, imaging, and legally enforceable work restrictions. Lacking psychiatric training, physical therapists are not in the best position to detect and intervene risk of work disability (which are primarily psychosocial) and physical therapy alone (as opposed to a multidisciplinary team) is not an appropriate treatment for chronic pain. This intervention embraced more complexity than the previous study by expanding the team to include appropriate expertise.
Implications for emergency medicine, physiatry, and policymakers
Efforts to council back pain patients to self-manage are appropriate, but will not prevent many from seeking emergency care. The emergency physician's duty to detect occasional dangerous spinal disorders that present to the ED, and to reassure all other patients, remains important. Interventions that provide some relief of acute symptoms, whether medication, physical modalities or even manual treatment may be useful, but physicians need to realize that they have no effect on short-term or long-term disability.[12],[13]
The challenge is to find a way for emergency physicians to not go further when they can do no more. First, they need to understand the futility of doing more in most cases. Diagnostic tests have risks and are rarely useful.[21] The “medically safe” MRI scan can double the rate of surgery without improving outcomes.[22] Abuse and misdirection of some medications are well known.[19] There is little an emergency physician should do for chronic pain, even when it presents with acute exacerbation. While an intramuscular shot of this or that might get a chronic pain sufferer out of the ED faster, this short-term pain vacation teaches chronic patients something counter to the real goal of knowing sustainable ways to function despite pain. The American College of Emergency Medicine, after review of opioid prescriptions for back pain, finds no strong support for their use, though admitting that the research was not conclusive.[20]
A publication using the pre-intervention cohort from this ED combined with data from two other ED explored some of the drivers of physician action.[23] It found that orders for advanced imaging were influenced by Caucasian race, age 50 years or older, warning signs, and radicular findings, whereas opioid prescriptions were predicted by severe pain and psychosocial “yellow flags” for disability risk.
EDs that hope to act competently regarding back pain need to take some steps that require a systemic approach. Their physicians need to realize that detection and management of disability, not just pathology, is their business. They must develop a discomfort with their skill set in managing back pain, and this must motivate them to seek more education. The system surrounding them needs to support good decision-making including a routinized intake process that detects the red and yellow flags. The physicians must come to realize the real harm of diagnostic testing in emergencies and weigh them against the remote risk of successful litigation when accepted standards are upheld. They need to build a partnership with expert physicians who care about long-term management.
By surrounding the ED in a complex network of knowledge, protocol, and partnership, this project showed substantial improvements in care. PM&R physicians, who are expert in nonsurgical spine care, are increasingly asked to work within their healthcare system to improve care and decrease cost. They can be highly effective if they reach out to help their colleagues in the ED. The process described here can be a reproducible, evidence-based platform for effective partnership.
Conclusions
This study demonstrates that emergency care of persons with back pain can be improved through a systematic approach that encompasses physician and patient education, process re-engineering, and longitudinal partnership with experts who can ensure competent care after the ED encounter. As healthcare systems are increasingly held responsible for overall health, their EDs will want to engage in the complex internal and external engagement needed to truly impact long-term quality and cost of care for this common, chronic, costly, and devastating disorder.
Financial support and sponsorship
This project was funded by the Center for Healthcare Research and Transformation.
Conflicts of interest
Dr. Haig is president of Haig Consulting and active emeritus professor of Physical Medicine and Rehabilitation at the University of Michigan.
References
| 1 | Dagenais S, Caro J, Haldeman S. A systematic review of low back pain cost of illness studies in the United States and internationally. Spine J 2008;8:8-20. |
| 2 | Corwell BN. The emergency department evaluation, management, and treatment of back pain. Emerg Med Clin North Am 2010;28:811-39. |
| 3 | Waterman BR, Belmont PJ Jr., Schoenfeld AJ. Low back pain in the United States: Incidence and risk factors for presentation in the emergency setting. Spine J 2012;12:63-70. |
| 4 | Sebert SL, Brick JE, Anderson K. Acute low back pain findings and management in an academic medical center. W V Med J 1998;94:202-4. |
| 5 | Chou R, Fu R, Carrino JA, Deyo RA. Imaging strategies for low-back pain: Systematic review and meta-analysis. Lancet 2009;373:463-72. |
| 6 | Katz JN, Gall V. Agency for health care policy and research clinical practice guideline for acute low back pain. Arthritis Care Res 1995;8:134-6. |
| 7 | Elam KC, Cherkin DC, Deyo RA. How emergency physicians approach low back pain: Choosing costly options. J Emerg Med 1995;13:143-50. |
| 8 | Reinus WR, Strome G, Zwemer FL Jr. Use of lumbosacral spine radiographs in a level II emergency department. AJR Am J Roentgenol 1998;170:443-7. |
| 9 | Jorgensen DJ. Fiscal analysis of emergency admissions for chronic back pain: A pilot study from a Maine hospital. Pain Med 2007;8:354-8. |
| 10 | French SD, Green S, Buchbinder R, Barnes H. Interventions for improving the appropriate use of imaging in people with musculoskeletal conditions. Cochrane Database Syst Rev 2010;CD006094. |
| 11 | Fox J, Haig AJ, Todey B, Challa S. The effect of required physiatrist consultation on surgery rates for back pain. Spine (Phila Pa 1976) 2013;38:E178-84. |
| 12 | Alvarez DJ, Chiodo AE, Graziano GP, Haig AJ, Harrison RV, Park P, et al. The Agency for Healthcare Research and Quality. Acute Low Back Pain (2010 Update). Ann Arbor, MI: Faculty Group Practice, University of Michigan Health System; 2010. Available from: http://www.guideline.gov. [Last accessed on 2016 Jul 01]. |
| 13 | Chou R, Qaseem A, Snow V, Casey D, Cross JT Jr., Shekelle P, et al. Diagnosis and treatment of low back pain: A joint clinical practice guideline from the American College of Physicians and the American Pain Society. Ann Intern Med 2007;147:478-91. |
| 14 | Lau PM, Chow DH, Pope MH. Early physiotherapy intervention in an accident and emergency department reduces pain and improves satisfaction for patients with acute low back pain: A randomised trial. Aust J Physiother 2008;54:243-9. |
| 15 | Webster BS, Courtney TK, Huang YH, Matz S, Christiani DC. Physicians' initial management of acute low back pain versus evidence-based guidelines. Influence of sciatica. J Gen Intern Med 2005;20:1132-5. |
| 16 | Indahl A, Velund L, Reikeraas O. Good prognosis for low back pain when left untampered. A randomized clinical trial. Spine (Phila Pa 1976) 1995;20:473-7. |
| 17 | Melloh M, Elfering A, Egli Presland C, Roeder C, Barz T, Rolli Salathé C, et al. Identification of prognostic factors for chronicity in patients with low back pain: A review of screening instruments. Int Orthop 2009;33:301-13. |
| 18 | Friedman BW, O'Mahony S, Mulvey L, Davitt M, Choi H, Xia S, et al. One-week and 3-month outcomes after an emergency department visit for undifferentiated musculoskeletal low back pain. Ann Emerg Med 2012;59:128-33.e3. |
| 19 | Zechnich AD, Hedges JR. Community-wide emergency department visits by patients suspected of drug-seeking behavior. Acad Emerg Med 1996;3:312-7. |
| 20 | Cantrill SV, Brown MD, Carlisle RJ, Delaney KA, Hays DP, Nelson LS, et al. Clinical policy: Critical issues in the prescribing of opioids for adult patients in the emergency department. Ann Emerg Med 2012;60:499-525. |
| 21 | Deyo RA. Imaging idolatry: The uneasy intersection of patient satisfaction, quality of care, and overuse. Arch Intern Med 2009;169:921-3. |
| 22 | Gilbert FJ, Grant AM, Gillan MG, Vale L, Scott NW, Campbell MK, et al. Does early imaging influence management and improve outcome in patients with low back pain? A pragmatic randomised controlled trial. Health Technol Assess 2004;8:iii, 1-131. |
| 23 | Kohns DJ, Haig AJ, Uren B, Thompson J, Muraglia KA, Loar S, et al. Clinical predictors of the medical interventions provided to patients with low back pain in the emergency department. J Back Musculoskelet Rehabil 2018;31:197-204. |
|
