Chronic Low Back Pain — Evidence-Based Care
Table of Contents
- Why This Article Exists
- Red Flags — When Back Pain Is an Emergency
- Inflammatory Back Pain — The AS Red Flag
- Imaging — Why MRI Often Makes Things Worse
- Most Chronic Back Pain Is Nociplastic, Not Structural
- The ACP 2017 Guideline in Plain English
- Movement Therapies — McKenzie, Motor Control, Feldenkrais
- Mind-Body Therapies — CBT, Yoga, Mindfulness, ACT
- Medications — What Actually Works
- Injections — Epidural Steroid and SI Joint
- Surgery — The SPORT Trial and Its Limits
- Pain Reprocessing Therapy — The Boulder Study
- The Biopsychosocial Model
- Building Your Own Care Plan
- Key Research Papers
- Research Papers
- Connections
Why This Article Exists
Low back pain is the single leading cause of years lived with disability worldwide. Roughly 80% of adults will have a significant episode in their lifetime, and about 20% of those episodes turn chronic — meaning pain persists longer than three months. If you have been in pain that long, you already know the drill: doctors order an MRI, the MRI shows "something," a procedure or surgery is offered, and either nothing changes or the pain returns in a different place six months later.
The scientific consensus has shifted dramatically in the past fifteen years, but medical practice has not caught up. The best evidence says most chronic low back pain is not caused by the structural findings on your MRI, that imaging in the absence of red flags makes outcomes worse, that first-line care is movement and mind-body therapy, and that opioids and most surgeries help far less than patients are told. This article walks through what the American College of Physicians, the Lancet series on back pain, and the major randomized trials actually recommend — and what you can do about it starting this week.
Red Flags — When Back Pain Is an Emergency
Before anything else: the minority of back pain cases where structure really is the problem and delay is dangerous. If any of these apply, stop reading and call your doctor or go to an emergency department today.
- Cauda equina syndrome — new loss of bladder or bowel control, numbness in the saddle area (inner thighs, perineum, buttocks), progressive weakness in both legs, or loss of sexual function. This is a surgical emergency within 24–48 hours. Delay causes permanent paralysis.
- Cancer red flags — known history of cancer, unexplained weight loss, pain that is worse at night and unrelieved by position change, pain lasting more than 4–6 weeks unresponsive to conservative care, age over 50 with new onset.
- Spinal infection — fever, IV drug use, recent bacteremia, immunosuppression, recent spine procedure, or severe night pain. Epidural abscess and discitis are missed routinely and devastating when delayed.
- Fracture — recent significant trauma, or minor trauma plus osteoporosis, chronic steroid use, or age over 70. Vertebral compression fractures are often missed on exam and require imaging.
- Progressive neurologic deficit — worsening foot drop, worsening leg weakness, or expanding numbness that you can track week by week.
- Inflammatory back pain pattern (see next section) — suggests ankylosing spondylitis or axial spondyloarthritis.
In the absence of these features, imaging does not help and often hurts. That is not a statement of medical laziness — it is a finding backed by decades of trials.
Inflammatory Back Pain — The AS Red Flag
The one inflammatory disease hiding inside the "chronic low back pain" population is axial spondyloarthritis, of which ankylosing spondylitis is the classic form. The average delay from symptom onset to AS diagnosis is still seven to ten years. Most of that delay comes from primary care physicians treating inflammatory back pain as ordinary mechanical back pain.
The pattern is unmistakable once you know it. Ask yourself:
- Did the pain start before age 45?
- Has it lasted more than three months?
- Is morning stiffness longer than one hour, easing only after you move around?
- Does the pain wake you in the second half of the night?
- Do you feel it in the buttocks, sometimes alternating sides?
- Does it improve with exercise and worsen with rest — the opposite of a mechanical disc or muscle problem?
- Do NSAIDs (ibuprofen, naproxen) work remarkably well?
If three or more of those apply, you need an MRI of the sacroiliac joints (not the lumbar spine) and an HLA-B27 blood test, ordered by a rheumatologist. The lab cost is under $150, the MRI is the definitive test, and catching axial spondyloarthritis early changes the trajectory of the disease. See HLA-B27 Explained for what that test means, and the arthritis overview for how the spondyloarthropathies fit with other inflammatory joint diseases.
Imaging — Why MRI Often Makes Things Worse
The single most important study for any chronic back pain patient to understand is Brinjikji et al. 2015 — a systematic review of 33 studies covering 3,110 people who had no back pain at all and received spinal MRIs anyway. The findings:
- Disc degeneration on MRI: 37% of asymptomatic 20-year-olds, rising to 96% of asymptomatic 80-year-olds.
- Disc bulges: 30% at age 20, 84% at age 80.
- Disc protrusions: 29% at age 20, 43% at age 80.
- Annular fissures: 19% at age 20, 29% at age 80.
Translation: almost every finding a radiologist writes on your MRI report — bulge, protrusion, degeneration, desiccation, facet arthropathy, foraminal narrowing — is present at similar rates in people without pain. Those findings are normal age-related changes, like gray hair and wrinkles, not pain generators.
What the studies also show is that getting an MRI for non-specific back pain makes outcomes worse. Patients who get early imaging have more surgery, more injections, more opioid prescriptions, higher costs, more work absence, and similar or worse pain scores compared with matched patients who skip imaging. The mechanism is straightforward: the MRI report plants the idea of a broken back, fear drives avoidance of movement, avoidance drives deconditioning, deconditioning drives more pain. The scan itself becomes a nocebo.
The ACP and every major guideline now say: no imaging for non-specific low back pain within the first six weeks, and no imaging after that unless red flags appear or a specific procedure hinges on the result. If you already have an MRI report full of alarming-sounding findings, the appropriate reaction is relief, not dread. Ask your clinician: "Given Brinjikji 2015, does any of this actually explain my pain?"
Most Chronic Back Pain Is Nociplastic, Not Structural
The International Association for the Study of Pain now recognizes three pain mechanisms: nociceptive (tissue damage — a cut, a sprain, arthritis), neuropathic (nerve damage — sciatica from a compressed root, diabetic neuropathy), and nociplastic (the nervous system itself has become the source). The third category covers fibromyalgia, chronic tension headache, irritable bowel syndrome, and — in most long-standing cases — chronic low back pain.
In nociplastic pain, the original injury has healed but the nervous system has learned to generate pain signals anyway. The brain's pain-processing networks become central sensitized: pain thresholds drop, non-painful touch starts feeling painful (allodynia), and the pain spreads beyond its original territory. This is not "psychological" pain — the pain is real, the brain activity is measurable on fMRI, and telling a patient "it's in your head" is both wrong and counterproductive. But the treatment is fundamentally different from treating a torn disc. See Central Sensitization and Nociplastic Pain and Pain Types Explained for the mechanistic detail.
Clues that your back pain has gone nociplastic:
- Pain has persisted long past the typical tissue-healing window (6–12 weeks).
- Pain location shifts — one day the left side, next week the right, next month the neck.
- Pain intensity correlates with stress, poor sleep, or emotional triggers more than with physical activity.
- You also have fatigue, brain fog, IBS, migraine, or other chronic pain syndromes.
- MRI shows findings no worse than your asymptomatic friends' findings.
- Light touch on the skin over the painful area feels unpleasant or burning.
If this sounds like you, read on — the good news is that nociplastic pain is often more reversible than structural pain, not less.
The ACP 2017 Guideline in Plain English
In 2017 the American College of Physicians published the most influential modern guideline on low back pain (Qaseem et al., Annals of Internal Medicine). Its central recommendation inverted fifty years of practice: for chronic low back pain, non-drug therapy comes first, drugs come second, and opioids come last if at all.
First-line therapies (strong evidence):
- Exercise — any kind you will actually do. Aerobic, strengthening, water-based, McKenzie extension, motor control. The specific type matters less than consistency.
- Multidisciplinary rehabilitation — a program that combines PT with CBT and graded activity.
- Acupuncture — modest but real effect sizes in meta-analyses.
- Mindfulness-based stress reduction (MBSR) — equivalent to CBT in head-to-head trials.
- Tai chi.
- Yoga — particularly Iyengar and Viniyoga styles studied in trials.
- Motor control exercise.
- Progressive relaxation.
- Electromyography biofeedback.
- Low-level laser therapy.
- Operant therapy.
- Cognitive behavioral therapy (CBT).
- Spinal manipulation.
Second-line (if first-line insufficient): NSAIDs (ibuprofen, naproxen) as the preferred drug. Duloxetine or tramadol as a next step. Acetaminophen is explicitly demoted — the Paracetamol for Low-Back Pain Study (PACE trial, 2014) showed it worked no better than placebo.
Last resort: opioids, and only "after a discussion of known risks and realistic benefits" and only when other options have failed. The implicit message — explicit in newer 2022 CDC guidance — is that opioids for chronic non-cancer back pain almost never make sense as long-term therapy. See Opioids for Chronic Non-Cancer Pain.
Movement Therapies — McKenzie, Motor Control, Feldenkrais
"Exercise" is the most under-prescribed, over-hedged therapy in all of medicine. Every meta-analysis says it works. The trick is finding the style you will stick with.
McKenzie Method (Mechanical Diagnosis and Therapy). A systematic approach built around repeated end-range movements — most famously prone press-ups (lying face-down, propping up on the elbows or hands). The therapist identifies the single movement that "centralizes" your pain (pulls it from the leg back toward the spine) and assigns it as homework, sometimes hundreds of reps a day. Works best for patients with a clear directional preference. Training takes time to find; look for a therapist with Cred-MDT or Dip-MDT credentials.
Motor control exercise. Teaches the deep stabilizing muscles (transverse abdominis, multifidus) to fire automatically before and during movement. Modest advantage over general exercise in trials, but the real benefit is reconnecting you with muscles you have stopped trusting. A well-trained PT can teach the core activation patterns in four to six sessions.
Feldenkrais and Alexander Technique. Movement-awareness disciplines that retrain habitual posture and tension patterns. Randomized trial evidence is modest but encouraging, particularly for people whose pain is tied to chronic guarding.
Yoga and tai chi. Multiple RCTs show durable benefit for chronic low back pain, comparable to conventional PT. Start with beginner Iyengar or gentle hatha — power yoga can re-injure a sensitized back.
Walking. The most underrated therapy. Thirty to forty-five minutes daily, outdoors if possible, builds aerobic fitness, releases endogenous opioids and endocannabinoids, and directly improves disc nutrition (the discs depend on movement for their fluid exchange).
The non-negotiable principle: movement is the medicine. Bed rest longer than 48 hours actively delays recovery. If you have been avoiding activity because you fear "making it worse," that fear itself is now part of the pain problem. See Sleep, Exercise, and Lifestyle for Chronic Pain.
Mind-Body Therapies — CBT, Yoga, Mindfulness, ACT
For chronic pain the brain is not a metaphor — it is the final common pathway where all pain is constructed. Interventions that change how the brain processes pain signals change the pain itself.
Cognitive behavioral therapy (CBT) for chronic pain. Eight to twelve sessions that target catastrophic thinking ("this pain will destroy my life"), fear-avoidance, activity pacing, and sleep. Effect sizes are modest but durable, and comparable to most pharmacotherapy. Many insurance plans now cover it.
Mindfulness-Based Stress Reduction (MBSR). The Cherkin 2016 JAMA trial compared MBSR, CBT, and usual care for chronic low back pain. MBSR and CBT both beat usual care, and the two were equivalent to each other. Eight weekly group sessions plus home practice. Community MBSR programs are often cheaper than therapy.
Acceptance and Commitment Therapy (ACT). Rather than trying to reduce pain, ACT teaches you to act according to your values even when pain is present — a subtle shift that paradoxically often reduces suffering and pain intensity.
Pain Neuroscience Education (PNE). Learning how pain actually works — that pain is produced by the brain, not the tissues; that your MRI findings are not the cause; that hurt does not equal harm — reliably reduces pain and fear in trials. Lorimer Moseley and David Butler's Explain Pain is the consumer book. See Pain Reprocessing Therapy and PNE.
Medications — What Actually Works
A blunt summary of the drug landscape for chronic non-specific back pain:
- NSAIDs (ibuprofen, naproxen, diclofenac, meloxicam): modest short-term benefit, roughly a 10-point improvement on a 100-point pain scale versus placebo. First-line drug. Watch stomach, kidney, and cardiovascular risk with long-term use.
- Acetaminophen: does not work for back pain. The PACE trial buried this one.
- Muscle relaxants (cyclobenzaprine, methocarbamol, tizanidine): mild short-term benefit, sedating. Reasonable for acute flares, not chronic daily use.
- Duloxetine: SNRI antidepressant with FDA approval for chronic musculoskeletal pain. Works independent of mood effect. Modest but real benefit.
- Gabapentin and pregabalin: disappointing in back pain trials. Useful for clear-cut radicular (sciatic) pain, weak for axial back pain, and now under scrutiny for overprescription and misuse. See Non-Opioid Medications.
- Tricyclics (amitriptyline, nortriptyline): low-dose, mostly for sleep and for pain with a neuropathic flavor.
- Tramadol: weak opioid plus SNRI action. Effective but still an opioid, with dependence potential often underestimated.
- Low-dose naltrexone (LDN): off-label, cheap, well-tolerated, with emerging evidence in nociplastic pain conditions.
- Opioids: never demonstrated long-term benefit for chronic back pain in any RCT. The SPACE trial (Krebs 2018, JAMA) randomized 240 patients with chronic back or knee pain to opioid versus non-opioid therapy for one year — non-opioid therapy won on pain and function. Do not start chronic opioids for back pain. See Opioids for Chronic Non-Cancer Pain.
Injections — Epidural Steroid and SI Joint
Epidural steroid injections (ESI) are among the most commonly performed procedures in pain medicine. The evidence is narrower than the practice suggests:
- For acute radiculopathy (sciatica from a disc) with leg pain worse than back pain, ESI produces a modest short-term benefit (2–6 weeks) on leg pain. Effect fades by 3 months.
- For axial (non-radicular) back pain, ESI does not reliably help.
- ESI does not change the rate of eventual surgery in most trials, though it may delay it.
- Risks are small but not zero: transient glucose elevation, rare dural puncture, very rare spinal cord injury (especially with transforaminal cervical injections).
Sacroiliac (SI) joint injections are both diagnostic and therapeutic. If a fluoroscopy-guided intra-articular injection of local anesthetic produces >75% pain relief for the duration of the anesthetic, the SI joint is the likely pain generator — a small but real subgroup. Therapeutic benefit from steroid varies; radiofrequency ablation of the sacral lateral branches offers longer relief in responders.
Medial branch blocks and facet radiofrequency ablation help a subset of patients with facet-mediated pain. Success rates reported in the literature are wide (30–80%) and depend heavily on strict patient selection with dual diagnostic blocks.
The honest framing: injections are a bridge, not a destination. They can buy weeks or months of reduced pain in which you do the movement and mind-body work that actually rewires the system. See Interventional Pain for the full menu.
Surgery — The SPORT Trial and Its Limits
The Spine Patient Outcomes Research Trial (SPORT), published in a series of JAMA and NEJM papers starting in 2006, is the most rigorous comparison of spinal surgery versus non-operative care ever performed. It covered three conditions: lumbar disc herniation, spinal stenosis, and degenerative spondylolisthesis. The headline findings:
- For disc herniation with sciatica: surgery produced faster pain relief than non-operative care in the first year, but by two years the two groups converged. Many patients in the non-operative arm improved without surgery; many who crossed over to surgery did well.
- For spinal stenosis: decompression surgery had a modest but durable advantage at 2–4 years for leg pain and walking capacity.
- For degenerative spondylolisthesis with stenosis: surgery had the largest advantage over non-operative care, but still roughly a third of non-operative patients did well.
- For non-specific axial back pain with degenerative changes but no clear nerve compression: fusion surgery has not shown durable benefit over intensive rehabilitation in any rigorous trial. This is the single most common reason chronic back pain patients are offered surgery, and it is the reason with the worst evidence.
What SPORT and the follow-on literature show is that spinal surgery is highly useful in a narrow set of indications (progressive neurologic deficit, severe refractory sciatica with matching imaging, clear stenosis with neurogenic claudication, cauda equina, fracture, tumor, infection) and unreliable to actively harmful outside those indications. "Failed back surgery syndrome" is common enough to have its own ICD code. If you are being offered fusion for axial back pain with MRI findings but no red flags and no clear neural compression, get a second opinion from a surgeon who does not financially benefit from operating, and ask what the trial evidence is for your exact indication.
Pain Reprocessing Therapy — The Boulder Study
The most important recent addition to the evidence base is Pain Reprocessing Therapy (PRT), tested in the Boulder Back Pain Study (Ashar et al., JAMA Psychiatry, 2022). The trial randomized 151 adults with chronic back pain (average 10 years of pain, average pain 4/10) to one of three arms: four weeks of PRT (nine one-hour sessions), placebo injection, or usual care.
Results:
- 66% of PRT patients were pain-free or nearly pain-free at the end of treatment, compared with 20% of placebo and 10% of usual care.
- Benefits were largely maintained at one-year follow-up.
- fMRI showed measurable changes in brain regions involved in pain processing.
PRT is built on the premise that chronic back pain is often a learned neural circuit — a false alarm the brain has gotten very good at producing. The therapy combines pain neuroscience education, somatic tracking (feeling the pain with curiosity rather than fear), and cognitive work on the fear-pain loop. It is not magic, it is not appropriate when red flags are present, and it does not work for everyone — but for the substantial fraction of chronic back pain patients whose pain is nociplastic, it is the most effective intervention yet tested. See Pain Reprocessing Therapy and PNE.
The Biopsychosocial Model
The old model of back pain was purely biomedical: something is broken in the spine, fix the thing, pain goes away. The old model fails for most chronic back pain. The biopsychosocial model, proposed by George Engel in 1977 and refined by pain researchers since, holds that chronic pain emerges from the interaction of three layers:
- Biological — tissue factors, inflammation, nerve sensitization, sleep, hormones.
- Psychological — catastrophizing, fear-avoidance, depression, anxiety, past trauma, attention and meaning.
- Social — work demands, relationships, disability incentives, cultural beliefs about pain, healthcare interactions.
Every one of those layers contributes measurably to chronic pain outcomes. Focusing only on the biological layer is why medicine has spent forty years ordering MRIs, doing injections, and watching chronic back pain get worse as a public health problem, not better. A proper care plan addresses all three layers simultaneously.
Building Your Own Care Plan
If you have had back pain longer than three months, with no red flags, and you want a starting point that matches the evidence:
- Rule out the dangerous stuff. A primary care visit to screen red flags. Inflammatory-back-pain questions to screen for axial spondyloarthritis. No imaging unless something is wrong with that screen.
- Find a PT who believes in the modern model — not one who gives you ultrasound and sends you home with three stretches. Ask if they practice motor control, McKenzie, or pain neuroscience education.
- Walk daily, starting from wherever you are, adding a few minutes a week.
- Pick one mind-body therapy and commit to eight weeks: MBSR class, CBT for chronic pain, yoga, or a PRT-oriented therapist.
- Read The Way Out by Alan Gordon or Explain Pain by Moseley and Butler. Books are cheap and the pain-neuroscience piece is non-negotiable.
- Fix sleep. Poor sleep amplifies pain the next day, measurably. See Sleep, Exercise, and Lifestyle for Chronic Pain.
- Treat NSAIDs as an adjunct, not a strategy. Use them for flares. Consider duloxetine if depressed or fatigued.
- Reserve injections for clear radiculopathy or a diagnostic SI-joint question.
- Be deeply skeptical of fusion offered for axial pain without red flags.
- Reassess at 12 weeks, honestly, with the same pain and function measures each time.
Key Research Papers
- Brinjikji W, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR. 2015.
- Qaseem A, et al. Noninvasive treatments for acute, subacute, and chronic low back pain: A clinical practice guideline from the American College of Physicians. Ann Intern Med. 2017.
- Ashar YK, et al. Effect of pain reprocessing therapy vs placebo and usual care for patients with chronic back pain: a randomized clinical trial (Boulder Back Pain Study). JAMA Psychiatry. 2022.
- Krebs EE, et al. Effect of opioid vs nonopioid medications on pain-related function in patients with chronic back pain or hip or knee osteoarthritis pain: The SPACE randomized clinical trial. JAMA. 2018.
- Cherkin DC, et al. Effect of mindfulness-based stress reduction vs cognitive behavioral therapy or usual care on back pain and functional limitations in adults with chronic low back pain. JAMA. 2016.
- Foster NE, et al. Prevention and treatment of low back pain: evidence, challenges, and promising directions. Lancet Low Back Pain Series. 2018.
- Weinstein JN, et al. Surgical vs nonoperative treatment for lumbar disk herniation: The Spine Patient Outcomes Research Trial (SPORT). NEJM. 2008.
- Williams CM, et al. Efficacy of paracetamol for acute low-back pain: a double-blind, randomised controlled trial (PACE). Lancet. 2014.
Research Papers
For further reading, the following PubMed topic searches return current peer-reviewed work on chronic low back pain assessment and management:
- Chronic low back pain clinical guidelines
- Low back pain and asymptomatic MRI findings
- Nociplastic pain and chronic back pain
- Pain reprocessing therapy
- SPORT trial lumbar surgery outcomes
- Epidural steroid injections for low back pain
- Inflammatory back pain and axial spondyloarthritis
- McKenzie method and mechanical diagnosis
- Yoga for chronic low back pain
- Mindfulness and chronic back pain
Connections
- Chronic Pain Overview
- Nociceptive vs Neuropathic vs Nociplastic Pain
- Central Sensitization and Nociplastic Pain
- Complex Regional Pain Syndrome
- Opioids for Chronic Non-Cancer Pain
- Non-Opioid Medications
- Interventional Pain
- Pain Reprocessing Therapy and PNE
- Sleep, Exercise, and Lifestyle for Chronic Pain
- Pain and Allergy
- Ankylosing Spondylitis
- HLA-B27 Explained
- Arthritis
- Ehlers-Danlos Syndrome