PROJECT SUMMARY
Low back pain represents a significant health care burden in the United States and accounts for nearly four
million emergency department (ED) visits per year. In nearly two thirds of these visits, an opioid medication is
administered or prescribed, making low back pain the most common reason for which opioids are prescribed.
Despite this aggressive medication-based approach, patient outcomes after an ED visit for back pain remain
poor: after three months, nearly half of all patients report persistent functional impairment, and one in five
patients report continued opioid use. Additionally, low back pain continues to be a major driver of unnecessary
diagnostic imaging. Nearly one in three ED visits for low back pain results in a plain radiograph, despite
multiple professional society guidelines advising against routine imaging.
ED-initiated physical therapy (ED-PT) is a promising new resource to improve patient care for low back pain.
A growing number of EDs now have dedicated physical therapists that can be consulted to evaluate and treat
patients through a combination of education, anticipatory guidance, and early mobilization and exercise. Our
preliminary data indicate that patients receiving ED-PT, compared to usual care, report more rapid functional
improvement, use fewer opioids, and receive less diagnostic imaging. However, these observational data are
limited by biases in treatment selection due to physician discretion in which patients receive ED-PT, as well as
other measured and unmeasured confounders.
To more rigorously evaluate the efficacy of ED-PT for acute low back pain, we propose (Aim 1) to develop a
revised model of ED-PT in which the physical therapist is “embedded” within the primary ED treatment team
to evaluate all patients with acute low back pain early in the overall treatment course: the Northwestern
Embedded Emergency Department Physical Therapy (NEED-PT) protocol. We will then (Aim 2) conduct a
physician-randomized trial of NEED-PT versus usual care (i.e., no physical therapy) in ED patients with acute
low back pain, comparing a primary outcome of patient-reported functioning and a secondary outcome of
opioid use at the primary endpoint of three months. Finally, we will (Aim 3) compare diagnostic imaging
utilization among ED back pain visits receiving NEED-PT versus usual care. We hypothesize that patients
receiving NEED-PT will experience greater improvement in functioning and lower use of opioids, and that ED
visits with NEED-PT will utilize less diagnostic imaging. The findings generated from this research have the
potential to meaningfully mitigate the current reliance on opioids for low back pain. As such, we believe this
proposal is both timely and responsive to the ARHQ Special Emphasis Notice on Health Services Research to
Address the Opioids Crisis (NOT-HS-18-015).
