Sequential and Comparative Evaluation of Pain Treatment Effectiveness Response

Chronic low back pain (cLBP) is common. The point prevalence of low back pain in the US is about 25%, and the majority of Americans will have at least one episode of low back pain in their lifetimes. Approximately 50% of all US Veterans have chronic pain, and cLBP is the most common type of pain in this population. Low back pain is the second most common condition leading patients to seek a physician's care. In addition, cLBP is costly. Healthcare expenditure for low back pain in the US is greater than $30 billion per year, and total expenditures including the cost of disability approach $100 billion per year. In fact, low back pain is the most common cause of work-related disability and is a major cause of service- connected disability amongst US Veterans. Making matters worse, the US is in the midst of a growing opioid abuse epidemic having its roots in the lack of understanding of how to effectively treat cLBP and other common forms of acute and chronic pain. Using a sequential randomized, pragmatic, 2-step comparative effectiveness study design, the main goal is to identify the optimal approach to cLBP treatment employing commonly recommended non-surgical, non-pharmacological options. Options for treatment in this trial were selected based on the VA's stepped-care model for the treatment of chronic pain, availability of treatments, controversies in current clinical practice, and the surveyed preferences of both Veterans with cLBP and VA healthcare providers. The implementation of study results has been kept closely in mind, and stakeholder input has been incorporated. The first step compares continued care and active monitoring arm (CCAM) to internet-based pain self-management (Pain EASE) and an enhanced physical therapy intervention that combines Pain EASE with tailored exercise and physical activity. The utility of tailored exercise requiring physical therapist guidance added to internet-based self-management has not been examined. A CCAM arm is included in this step to definitively assess the effectiveness of these initial treatment options. Patients failing to achieve clinically significant reductions in pain interference (a functional outcome) in Step 1 and those desiring additional treatment will be randomized in Step 2 to yoga, spinal manipulation therapy (SMT), or therapist-delivered cognitive behavioral therapy (CBT). These options appear in consensus guidelines although little information is available to help patients and providers select the most effective option. Each option has a distinct theoretical basis for effectiveness with yoga described as a mind-body therapy, SMT as a technique to adjust the structural relationships of the spine, and CBT as a psychological or behavioral approach. While the literature suggests approximate clinical equipoise between these treatments, costs, side effects, access to specific options and patient/provider acceptance may differ significantly. Participants proceeding to randomization in Step 2 will be allowed to exclude up to one of the three Step 2 treatments based on their preferences. The investigators' primary hypothesis corresponding to the first treatment step is that an enhanced physical therapy intervention that combines pain self-management education with a tailored exercise program will reduce pain interference greater than internet-based pain self-management alone or CCAM in Veterans with cLBP. For the second step, the study has been powered to detect clinically meaningful pairwise differences among the three treatments. The primary outcome is change in pain interference at 3 months, measured using the Brief Pain Inventory (BPI) pain interference subscale. This is a widely accepted functional outcome in musculoskeletal pain trials and emphasizes an endpoint important to patients, providers and healthcare management systems. Study participants will be followed for one year after initiation of their final study treatments to assess the durability of treatment effects. The investigators will leverage the power of this large study and maximize its impact by incorporating additional outcome measures as recommended by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) consortium and other pain trials consensus groups. These secondary outcomes include pain severity, physical functioning, depression, anxiety, fatigue, sleep, global impression of change, and quality of life. The investigators will evaluate the impact of patient characteristics, treatment preferences and expectations on study outcomes as these variables have been identified in smaller studies to modify treatment response. To derive information rapidly translatable to changes in VA care, the investigators will collect information critical for implementation of the treatment strategies under study. Key implementation factors will include treatment fidelity, treatment adherence, patient acceptance, provider acceptance, logistical feasibility, and resource requirements. Finally, a carefully designed costs and downstream budget impact aim will provide additional practical information for clinical managers and policy makers related to non-pharmacological treatments for cLBP. The study will involve diverse VA centers with respect to geographical region, racial characteristics of the population served and facility size. The study plans to randomize 2529 patients across 20 centers. Preliminary site surveys indicate a high level of enthusiasm for the project. In addition, a query of the VA's Corporate Data Warehouse (CDW) identified more than 850,000 potential cLBP study subjects receiving regular care through VA. Twenty-five VA medical centers have more than 10,000 potentially eligible patients underscoring the high prevalence of cLBP. The investigators' pragmatic trial design will incorporate broad eligibility criteria. Chronic low back pain is an enormous problem for the VA, the United States and many other countries. This study will provide definitive information concerning the effectiveness, costs, acceptability, and implementation of several commonly used, patient-preferred, non-pharmacological treatment options. All the selected treatment options carry relatively low levels of risk, are guideline congruent, and are consistent with stepped-care models of healthcare delivery used by the VA and other healthcare organizations. The impact on VA healthcare is expected to be large and immediate.