OPTIMIZE in Annals: Physical Therapy First for Chronic Low Back Pain, and What That Actually Means for Conservative Care
Fritz and colleagues randomized 749 adults with chronic low back pain across three health systems using a SMART design that adapted second-stage treatment based on early response. Patients started with either evidence-based medical care or physical therapy. Those who did not respond were re-randomized to add mindfulness or to switch the other arm in. PT first beat medical care first. Among nonresponders, second-stage choices did not separate. The trial reads as a vote for starting conservative, not as a vote for one specific brand of conservative.
Key Points
- OPTIMIZE enrolled 749 adults with chronic low back pain (12 or more weeks of symptoms) across three United States health systems from March 2019 through September 2023. Follow-up ran 52 weeks.
- Design was SMART (sequential multiple-assignment randomized trial): patients were randomized at study entry to first-stage treatment, then nonresponders were re-randomized at 10 weeks to a second-stage strategy. This is uncommon in low back pain research and lets the trial answer two questions at once.
- First-stage randomization: evidence-based medical care versus physical therapy. PT first produced meaningfully better disability and pain outcomes than medical care first.
- Second-stage randomization (for nonresponders): add mindfulness-based stress reduction, or switch to the alternate first-stage arm. Neither strategy outperformed the other at 52 weeks.
- Author conclusion in plain language: chronic LBP patients may benefit from physical therapy as a first-line treatment. Among those who do not respond, the specific next move appears to matter less than the fact that the patient is still inside conservative care.
- The Vanguard Chiropractic read: this is a vote for "conservative care first," not a vote that PT is the only acceptable on-ramp. Chiropractic care sits in the same conservative-care category alongside PT in every major clinical guideline. The trial validates the broader principle.
If you have had back pain for more than three months, what is the first move that gives you the best shot at less disability a year from now? A multisite team led by researchers at the University of Utah enrolled 749 adults with chronic low back pain and ran one of the more carefully designed trials this question has seen.
Everyone started in one of two paths. Half were sent into physical therapy. The other half got what the trial called evidence-based medical care, which is the package a primary care doctor would assemble: a structured visit, advice on staying active, simple medication options, and standard follow-up. At ten weeks, the patients who had not improved were sent back into the trial for round two. Some had mindfulness-based stress reduction added on top of what they were already doing. Some were switched into the other first-stage arm (so PT patients moved into medical care, and medical care patients moved into PT).
One year out, the patients who started with physical therapy were doing better than the patients who started with medical care. Among the patients who did not respond to either starting point, the second-stage choice did not seem to matter. Mindfulness was not better than switching. Switching was not better than mindfulness.
The headline most outlets are running is "PT first for chronic back pain." That is fair. The deeper headline is that being inside an active, conservative care plan beat sitting in a passive one, and that once you are inside that active plan, the specific next move matters less than the fact that you are still inside it. Chiropractic care lives in the same conservative-care category as physical therapy in every major guideline, and the trial says less about which brand of conservative wins than about how badly the default of "wait and see plus a pill" loses.
Why This Paper Matters
Chronic low back pain is the most expensive musculoskeletal condition in the United States. Standard primary care does poorly with it. The 2017 American College of Physicians guideline (Qaseem et al., Annals of Internal Medicine 2017) ranked nonpharmacologic care, including spinal manipulation, exercise, and mindfulness, ahead of medication for first-line treatment. Insurance pathways have been slow to follow. Most patients still hit a primary care visit, a prescription, and a wait-and-see strategy before any conservative provider sees them.
OPTIMIZE tested a simple counterproposal. Send the patient directly to physical therapy. Skip the wait. See what happens at one year. The trial then layered a second question on top of the first by using a SMART design: when the patient does not respond to the first move, what is the best next move? This is the question every clinician actually has to answer in practice, and it is almost never the question randomized trials are powered for.
The result is one of the cleanest pieces of conservative-care-first evidence published in a top medical journal in the last five years. It pairs naturally with the PACBACK trial in JAMA, which tested the same principle (start with structured conservative care) in a slightly earlier-onset population.
What a SMART Trial Actually Is
A standard randomized trial answers one question. You give half the patients drug A and half the patients drug B, and you measure who does better. The real world is messier than that. Most chronic conditions involve a first treatment, a check-in, and a second treatment if the first one does not work. Almost no trial is designed to test the whole sequence.
SMART (sequential multiple-assignment randomized trial) designs solve that gap. Patients are randomized at the start of treatment AND at a planned re-check point if they have not responded. The output is not just "which first treatment wins," but "which sequence of treatments produces the best outcome." It is a closer mirror of clinical practice than a two-arm trial.
For OPTIMIZE, the structure was:
Stage 1: All 749 patients randomized to evidence-based medical care (EBMC) or physical therapy (PT). Each arm ran for about 10 weeks.
Re-assessment at 10 weeks: Patients with a clinically meaningful improvement in disability (per the Oswestry Disability Index) continued the same care. Patients who had not responded entered the second randomization.
Stage 2 (nonresponders only): Re-randomized to add mindfulness-based stress reduction (MBSR) on top of current care, or to switch into the alternate stage-1 arm.
The 52-week outcome compared disability and pain across the full embedded sequences (PT-then-MBSR, PT-then-switch-to-EBMC, EBMC-then-MBSR, EBMC-then-switch-to-PT, and the responder paths). The headline comparison is the stage-1 winner. The interesting comparison is among the nonresponder sequences.
What They Found
At 52 weeks, patients in the PT-first arm showed clinically meaningful reductions in disability compared to patients in the EBMC-first arm. Pain scores moved in the same direction. The effect size sits in the range where guideline bodies have historically treated a finding as actionable for first-line care.
Among nonresponders to stage 1, the second-stage randomization produced no statistically significant separation. Mindfulness added on did not outperform switching arms. Switching arms did not outperform mindfulness added on. The author conclusion was direct: "Among nonresponders, there were no differences in second-stage treatment with mindfulness or switching."
Read that finding carefully. Two things follow.
First, the stage-1 choice matters. Patients who started with PT did better at one year than patients who started with EBMC. The clinical-policy implication is that the front door of chronic LBP management probably should not be a primary-care visit followed by a pill and a wait. It should be an active, hands-on, structured conservative care contact.
Second, when the first move does not work, the literature does not yet tell us what the best second move is. The trial tested two reasonable candidates and they both performed about the same. That is a useful piece of clinical humility. The patient who is not getting better at 10 weeks is a real clinical population, and the best move for them is probably "keep trying something inside conservative care" rather than "the algorithm says do X."
Where Chiropractic Sits in This Picture
OPTIMIZE did not include a chiropractic arm. The trial tested PT versus EBMC at stage 1, and MBSR versus switching at stage 2. A reader could finish the abstract and walk away thinking the paper says "PT, not chiropractic." That is not what the paper says.
Here is what it actually says, read in the context of the broader guideline literature.
Every major clinical guideline for low back pain groups chiropractic spinal manipulation and physical therapy together inside the same first-line category. The ACP guideline lists spinal manipulation alongside exercise, heat, and acupuncture as appropriate first-line care. The NICE guideline (NG59) in the United Kingdom does the same. The U.S. Department of Veterans Affairs guideline does the same. The distinction the literature consistently draws is between active conservative care (any of the above) and passive medical management (medication, rest, observation). OPTIMIZE is a test of the second of those two categories versus the first. PT was the first-line arm because the trial had to pick one operationalizable provider, not because the literature says PT is the only acceptable on-ramp.
The honest read of OPTIMIZE for chiropractic patients is this: starting with active conservative care produces better outcomes than starting with passive medical care. Active conservative care includes physical therapy, chiropractic, and exercise-based rehab. The trial's design picked one of those to test. The principle applies more broadly.
This is consistent with the dos Santos 2026 review on manual therapy plus exercise, which found that manual therapy and exercise together produced larger effects than either alone. It is consistent with the Liang 2024 Bayesian network meta-analysis, which ranked active modalities at the top of the dose-response hierarchy. It is consistent with the 2026 Cochrane review placing spinal manipulation in the same effectiveness band as other conservative options. The literature has been pointing in this direction for ten years. OPTIMIZE is the latest data point, and it is a high-quality one.
The Mechanism Question
Why does active conservative care outperform passive medical care for chronic low back pain? The mechanisms are not mysterious.
Chronic low back pain is associated with measurable changes in motor control, fear-avoidance behavior, deconditioning, and central pain processing. None of these respond to a prescription. They respond to graded loading, exposure to feared movement, restored joint mobility, and consistent contact with a clinician who treats the patient as someone who is going to get better.
Physical therapy delivers most of these elements. Chiropractic care delivers most of these elements with a different emphasis on the joint-mobility piece and a different visit format. Both arms of the first-line literature share the active ingredients. Wait-and-see medical care delivers almost none of them. The OPTIMIZE result is what you would predict if you started with that mechanism map and asked which arm would win.
The second-stage null result also makes mechanistic sense. A patient who has not responded to one form of active conservative care has not exhausted the active-care category. They have exhausted that specific dose, with that specific provider, at that specific time. Adding a stress-reduction component on top, or switching providers, are both reasonable second moves. Neither is structurally superior. The principle is that the patient stays inside active care, with adjusted inputs, rather than dropping out into passive management.
What I Tell My Patients
Most of my chronic low back pain patients arrive at Vanguard Chiropractic having already cycled through some version of the passive medical management arm of this trial. Primary care visit. Prescription. Maybe an MRI. Wait and see. By the time they walk in, they have usually been at the problem for months or years, and the wait-and-see arm has not delivered.
When I read OPTIMIZE in that context, the clinical message is straightforward. The first move on chronic low back pain should be active, hands-on, conservative care. That is what I deliver. It is what a physical therapist delivers. It is what a movement-focused exercise program delivers. The data say start there.
For chronic radicular and recurrent presentations, I recommend ongoing weekly care for at least twelve weeks after the initial acute phase resolves, then a taper. This is not maintenance for its own sake. Each flare of nerve-root irritation compounds tissue irritation and central sensitization patterns, and the responsible approach is to prevent flares before they happen rather than chase them after. The OPTIMIZE second-stage finding (that the patient who has not responded should stay inside conservative care, with adjusted inputs) is consistent with the same principle: stay in the active-care loop until the system has stabilized.
I do not run a six-session structured psychoeducation protocol the way the PT arm of OPTIMIZE did. My visit format is shorter and the price point is lower. What I can do, and do at every visit, is layer brief active-care elements into a $40 adjustment encounter: a movement to take home, a fear-avoidance reframe when the patient says "I am afraid to bend," a graded-loading suggestion, a return instruction. None of these match a one-hour PT session in isolation. Across multiple visits, they build the same scaffolding.
Limitations and Caveats
OPTIMIZE did not include a chiropractic arm. Direct extrapolation to chiropractic care relies on the parallel guideline placement of chiropractic and PT inside the same first-line category, not on a head-to-head test. A future SMART trial that tested PT versus chiropractic at stage 1 would be the cleaner comparison.
The evidence-based medical care arm reflects how primary care actually treats chronic LBP, which is a low bar. Beating EBMC is meaningful, but it is not the same as beating an optimized, integrated primary care pathway that aggressively refers to conservative providers. The trial does not test that comparator.
The second-stage null result on mindfulness versus switching is hypothesis-generating, not definitive. The trial may have been underpowered to detect a small but real difference between these strategies. A bigger trial may yet separate them.
Adherence to the assigned care was monitored but is not perfect in any pragmatic trial. Real-world adherence to either PT or EBMC over 10 weeks varies by patient, and the intention-to-treat analysis carries the structural noise that comes with that.
Three health systems is more external validity than a single-center trial provides, but the United States health-system context still differs from the international guideline environment. Findings extrapolate most cleanly to insurance-pathway populations and less cleanly to cash-pay or uninsured populations where the access pattern is different.
The trial cannot distinguish "PT first works because PT is uniquely effective" from "PT first works because any structured active care delivered at the front door beats wait-and-see." The second interpretation is what the broader guideline literature supports. The trial design cannot adjudicate between the two.
Clinical Context: Conservative Care First, Not Conservative-Brand-A First
The risk in how this paper will be reported is the brand-versus-principle confusion. Outlets will write "PT first for chronic back pain." Policy people will hear "PT, not chiropractic." That is not what the trial says. It says active conservative care, delivered as the first move, beats passive medical management. PT was the specific operational arm tested. The principle is broader.
For a Tulsa patient sitting at their kitchen table trying to decide what to do about back pain that has not gone away in three months, the practical implication is the same whether they read OPTIMIZE, PACBACK, the dos Santos review, or the Cochrane verdict. Start with active conservative care. Pick the provider you can access, afford, and stick with. A physical therapist works. A chiropractor focused on back pain works. A structured exercise program works. The wait-and-see strategy does not.
For my walk-in patients in Brookside, the access piece is the part that matters most. $40 per visit, no appointment required, open six days a week. That is not a research-trial intervention. It is a practical answer to the "pick the provider you can access, afford, and stick with" sentence in the paragraph above.
749 adults with chronic low back pain across three health systems, a SMART design that adapted treatment based on early response, and a clean first-stage finding: starting with physical therapy beat starting with usual medical care at one year. Among nonresponders, the specific second move did not separate. The story the trial actually tells is "active conservative care first," not "PT, not chiropractic." Chiropractic, PT, and structured exercise all live in the same first-line guideline category. The trial reinforces what the literature has been pointing at for a decade: wait-and-see is the loser, and any responsible front door of active care is the winner. For a patient with chronic LBP, the best move is to start, not to wait.
Editorial Disclosure
The Forward Review is editorial commentary, not peer reviewed. Written by Jason Forward, DC, a licensed chiropractor at Vanguard Chiropractic in Tulsa, Oklahoma. Commentary reflects clinical interpretation of published research and does not represent the views of the original study authors, the journal Annals of Internal Medicine, or any institution.
Conflict of interest: Dr. Forward is a practicing chiropractor whose clinical work includes spinal manipulation as part of a multimodal approach to musculoskeletal care. No financial relationships with the cited authors, Annals of Internal Medicine, or any party to this work.
Scope of practice note: The OPTIMIZE trial studied physical therapy delivered by licensed PTs and medical care delivered by primary care clinicians. This commentary discusses published research and does not constitute individual treatment prescription. Patients with new or worsening back pain should be evaluated by a qualified clinician; red flags including unexplained weight loss, fever, progressive neurological deficit, bowel or bladder changes, or saddle anesthesia require urgent medical evaluation.
Corrections policy: Errors are corrected with a dated correction notice appended to the article. Original text remains accessible via revision history.
Level of evidence: Sequential multiple-assignment randomized trial, n=749, three United States health systems, 52-week follow-up. Published in Annals of Internal Medicine, April 21, 2026.
Continue here: Why This Paper Matters
- Dual-Task Core Stability Training for Chronic Low Back PainForward Rev. 2026;1(2)
- When Back Pain and Low Mood Feed Each Other: 2026 Evidence on MovementForward Rev. 2026;1(2)
- Why Treating the Mid-Back Helps the Shoulder: 2026 Thoracic Manual Therapy EvidenceForward Rev. 2026;1(2)
- Nonsurgical Treatment for Lumbar Spinal Stenosis: 2026 Evidence RankedForward Rev. 2026;1(2)
View Vol. I, No. i and the full archive