Acceptance and Commitment Therapy for Chronic Pain: What the Evidence Actually Shows

What this video covers

• Why pain that outlasts tissue healing is best explained by changes in central pain processing rather than by tissue damage alone — and why that model is still a working framework drawn from imaging and animal studies, not proven human physiology
• How pain-related fear and avoidance lead to deconditioning and lost activity, and why disability correlates only loosely with what shows on a scan
• What ACT actually does: acceptance of sensation, defusion from catastrophic thoughts, and committed action toward what you value
• The real numbers from 33 randomized trials in 2,293 patients, including small-to-moderate gains in function and mood that shrink as follow-up lengthens
• Why the 2020 Cochrane review still rates the ACT evidence very low certainty against active comparators, and what that means for you
• Honest risks and limits: a short-term rise in distress when you stop avoiding, a flare in pain or fatigue as activity is graded upward, attrition near 20 percent in the fibromyalgia trials, adverse events poorly reported across the whole ACT literature, and no change to structural pathology
• MEDICAL DISCLAIMER: This content is for educational purposes only and is not medical advice. It does not substitute for professional diagnosis or treatment. Always consult a licensed healthcare provider regarding your condition. Viewing this video does not establish a doctor-patient relationship.

You have stopped taking the stairs. You sit in the one chair that does not make it worse. The MRI was two years ago, the first injection helped, and the fourth did not. Somewhere in there your world contracted to the recliner and the kitchen.

Here is what is missing from your chart. There is a column for morphine equivalents. There is no column for how long you can stand, how far you walked this week, or how many things you said no to. This article explains what that missing column measures, why it changes how pain feels, and where Acceptance and Commitment Therapy (ACT) — a behavioral treatment with real but modest trial evidence — belongs in the sequence of care. It is an explanation, not a sales pitch: the honest version of this evidence includes a Cochrane review that found no statistically significant benefit when ACT was measured against another active treatment.

The finding nobody wrote down: fear and avoidance

Alongside whatever your disc or joint is doing, a behavior set has been accumulating — every movement, outing, and obligation you deleted because you associate it with a flare.

Fear and avoidance are among the best-documented correlates of disability in chronic pain. The prospective data are consistent that they track disability better than imaging does. Be exact about what that means: a strong association with a plausible mechanism, not a proven cause in any individual patient. Nobody measured yours — not because it is unmeasurable, since the questionnaire takes about four minutes, but because no procedure attaches to the answer.

The mechanism, and the edge of it

Before anyone asks you to do behavioral work, you are owed the mechanism — and the edges of it, because here the confident version and the honest version are not the same version.

Acute pain is a report: nociceptors (the nerve endings that detect tissue threat) fire, tissue heals, the report stops. Pain that persists past that window looks different. The leading model describes amplified signaling in the dorsal horn of the spinal cord, weakened descending inhibition from the periaqueductal gray and rostral ventromedial medulla (brainstem regions that normally turn pain signals down), and threat learning across the amygdala, insula, and cingulate cortex. In that model the nervous system is not reporting damage. It is predicting it.

That account is drawn from human imaging and animal work. It is the leading framework. It is not confirmed human physiology, and reciting it as fact is overselling. Proposed brain-network explanations, including default-mode-network changes, remain preliminary. What is not in dispute is narrower and more useful: disability tracks imaging poorly.

Why the loop makes the pain feel like damage

This is the part that explains the lived experience rather than the diagram.

A sensation arrives. It is read as damage. You guard, and you cancel. Guarding produces deconditioning. Deconditioning shrinks tolerance, so the next ordinary movement hurts more — and the original interpretation is confirmed. The loop tightens on every revolution:

sensation → interpreted as damage → guarding → deconditioning → lower tolerance → confirmation

In this model, avoidance is an excellent short-term analgesic. The cohort data suggest it is a costly long-term one. That is why the stairs feel more dangerous this year than last even when the scan is unchanged, and why the flare that follows a good day feels like proof you were right to stop. It is not proof. It is deconditioning meeting an interpretation that was never tested.

What standard care misses — and what it must not skip

None of the above earns you a psychological label. You do not get to invoke central processing until you have earned it, and the way you earn it is by finishing the diagnosis.

Red flags come first. New weakness, a change in bowel or bladder control, fever, unexplained weight loss, night pain, or a history of cancer means imaging and laboratory work today, not psychology. The differential is resolved with anatomy, not adjectives.

The structural question deserves a real answer. In the video, Dr. Padda notes that no history item and no examination maneuver reliably identifies the facet joints as the pain generator — Maas and colleagues, European Journal of Pain, 2017, put the sensitivity of physical examination anywhere from eleven to one hundred percent. The answer is a low-volume medial branch block under live fluoroscopy, with the percentage of pain relieved during the anesthetic window as the only readout. Even then, false positives on a single block run twenty-seven to forty-seven percent, so one positive block is a hypothesis, not a diagnosis.

Burning numbness in both feet is a polyneuropathy until proven otherwise. The neurology practice parameter names three highest-yield screening tests: a blood glucose; a B12 with its metabolite methylmalonic acid, because a normal B12 can be functionally deficient; and a serum protein immunofixation electrophoresis to catch a paraprotein. If that routine glucose is not diabetic, the same parameter says a glucose tolerance test may be considered — a deliberately weaker recommendation than the first three, and worth naming as weaker. Dr. Padda adds a fasting insulin; that addition is his practice preference, not a guideline recommendation.

What standard care misses is therefore two-sided. Symptom-control medication and well-chosen injections do what they do well — they reduce a signal, and a properly targeted diagnostic block answers a question no questionnaire can answer. What neither does is measure or change the behavior set. A treatable driver hiding under a psychological label is the failure in one direction; an unmeasured, unaddressed avoidance pattern is the failure in the other.

What ACT actually is, and where it goes in the order

The order is the whole argument. Behavioral work goes third, never first. A program handed to a patient whose diagnosis is unfinished is not treatment; it is a way of closing the file.

1. First, the structural question gets answered and named, with a block that answers a question rather than filling a slot.
2. Second, the metabolic question, because a starved nerve does not respond to psychological flexibility. Be exact about what is proven here: ketogenic nutrition, specialized pro-resolving mediators, and photobiomodulation between 660 and 850 nanometers are investigational adjuncts, not standard of care.
3. Third — and only third — behavioral work.

The full ACT model has six processes. In pain programs, four do most of the work:

• Acceptance — you stop fighting a sensation you cannot switch off. This is not resignation to a permanent sentence; it is dropping a struggle that costs energy and wins nothing.
• Defusion — you watch the thought that your spine is disintegrating instead of obeying it.
• Values — you name what pain has taken: the grandchild, the workbench, the mile walk.
• Committed action — you graduate activity toward those things, on your schedule.

A licensed, pain-trained psychologist delivers this, commonly across six to twelve sessions, and never inside a procedure visit. Candidates are adults past three months of pain — three months, not a year — whose workup has cleared deficit, infection, fracture, and malignancy, and whose lives have narrowed around avoidance. The higher your avoidance, the more there is here for you. The trials measured benefit over three to twelve months; that is the honest horizon. If your pain is six weeks old, or the workup is unfinished, you are not a candidate today.

What the trials actually show

The largest synthesis is a three-level meta-analysis by Lai and colleagues (2023): 33 randomized trials, 2,293 patients. Post-treatment Hedges g was 0.44 for pain intensity, 0.59 for physical function, and 0.43 for both depression and anxiety. Real, and modest.

Two qualifiers matter. The first is time. Longer follow-up correlated with smaller effects for pain intensity and physical function: pain intensity fell from 0.44 to 0.34, physical function from 0.59 to 0.56. The mood estimates drifted less — depression was unchanged at 0.43, anxiety eased from 0.43 to 0.35 — but “drifted less” is not “held”. Anxiety came down by roughly as much as pain intensity did; what the analysis actually tested and found was erosion of the pain and function benefit over time. The second qualifier is delivery: face-to-face ACT produced significantly larger effects than internet-delivered ACT on the physical outcomes. The app is not the same product.

For fibromyalgia specifically, Eastwood and Godfrey (2024) pooled six trials and 384 patients, predominantly women. Pain acceptance improved substantially (SMD 1.05; 95% CI 0.61 to 1.49) — with the honest caveat that this is a self-report questionnaire score in unblinded trials, which caps how much weight it can carry. Fibromyalgia impact improved too, but imprecisely — SMD −1.05, with a confidence interval of −2.02 to −0.09 that very nearly touches no effect.

Now the hardest number, and the one that separates an honest article from a brochure. The 2020 Cochrane review by Williams and colleagues asked the harder question: not whether ACT beats a waiting list, but whether it beats an active comparator — another real treatment delivered with equal attention. That comparison is small. Five studies, 443 patients. Pain (SMD −0.54; 95% CI −1.20 to 0.11), disability (−1.51; −3.05 to 0.03), and distress (−0.61; −1.30 to 0.07) all favored ACT numerically, and not one reached statistical significance. Every confidence interval crossed zero. The evidence was graded very low certainty, and the review’s authors say plainly that they are very uncertain about benefit or the lack of it.

Read that correctly. Compared with doing nothing, ACT helps. Compared with another active treatment delivered with equal attention, it has not been shown to win. Very low certainty does not mean it fails; it means the trials were too small, too unblinded, and too self-reported to tell us either way. This is contested territory, and you should be told so.

The risks, and what ACT does not do

The risks are not zero.

• Distress rises before it falls. When you stop avoiding, grief or old trauma can surface. If you carry a post-traumatic stress diagnosis, trauma-focused treatment belongs alongside or before this work, because facing avoided experiences can briefly intensify trauma symptoms. If anxiety or post-traumatic stress is part of your picture, the clinical detail lives on our page about anxiety, PTSD, and chronic pain.
• Graded activity often brings a short-term flare in pain or fatigue as you step activity upward.
• Mood can worsen, and that warrants a phone call to your clinician rather than waiting it out.
• People drop out. Attrition in the fibromyalgia trials ran near twenty percent, and stayed under that mark in four of the six studies. At the high end that is about one patient in five who did not finish, and the safe assumption is that they did worse than the people who stayed.
• Safety is not established. The fibromyalgia review reported no adverse events attributable to ACT, but the Cochrane review rates adverse-event reporting across this whole literature as inadequate. Absence of reported harm is not the same as demonstrated safety.
• ACT changes nothing structural. A torn annulus is still torn.

What this means for your evaluation

If nobody has handed you a diagnosis you can name, that is the first thing to fix — not the last.

A course that goes well looks less dramatic than patients expect. The following is an illustrative composite assembled from many patients with this presentation, not one person’s chart, and the numbers in it show the shape of a good course rather than an average or a schedule to hold yourself to. An adult with axial low back pain past four years. Two surgical consultations, both declined. Medial branch blocks that were diagnostically informative, with real but finite relief. A nearly empty activity log. The first six weeks of behavioral work were worse, not better. The values work surfaced grief. There was a flare in week five that felt like proof the whole thing was a mistake. It was not proof. It was week five. In a course that goes well, the graded walk is measured in minutes by around the fourth month and in fractions of a mile by the ninth. These figures are illustrative rather than trial results. A meaningful number of patients never get there, and some of the gain that is made attenuates over time. Individual results vary.

That is not a cure, and nobody should offer you one. It is function, returned, purchased with work — layered on top of an accurate structural and metabolic diagnosis, not instead of one. If anxiety, post-traumatic stress, or low mood is part of what you are carrying into that evaluation, our page on how anxiety, PTSD, and chronic pain interact covers that side of the workup in clinical detail.

Frequently asked questions

Key takeaways

• Alongside the structural problem, an avoidance pattern accumulates — and fear and avoidance track disability better than imaging does, though the association is not proof of cause in any one patient.
• The central-processing account of persistent pain is the leading framework drawn from imaging and animal work, not confirmed human physiology.
• ACT’s evidence is real but modest: small-to-moderate effects across 33 trials, with the pain and function benefit eroding at follow-up, and no statistically significant benefit against an active comparator in the 2020 Cochrane review, graded very low certainty.
• Risks include a rise in distress, surfacing grief or trauma, activity-related flares, and dropout approaching one in five in the fibromyalgia trials; adverse-event reporting is inadequate, so safety is not established.
• Behavioral work goes third — after the structural question and the metabolic question are answered. Get the diagnosis first, then decide whether you want to do the work.

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed July 2026.

This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary. Do not start, stop, or change any medication without consulting your physician. To be evaluated, request an appointment at painmd.tv, call (314) 481-5000, or text (314) 886-5902.

References

1. Lai L, Liu Y, McCracken LM, Li Y, Ren Z. The efficacy of acceptance and commitment therapy for chronic pain: A three-level meta-analysis and a trial sequential analysis of randomized controlled trials. Behav Res Ther. 2023;165:104308. PMID 37043967. PubMed
2. Williams ACC, Fisher E, Hearn L, Eccleston C. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database Syst Rev. 2020;8:CD007407. PMID 32794606. PubMed
3. Eastwood F, Godfrey E. The efficacy, acceptability and safety of acceptance and commitment therapy for fibromyalgia – a systematic review and meta-analysis. Br J Pain. 2024;18(3):243-256. PMID 38751564. PubMed

Dr. Gurpreet Singh Padda, MD, MBA, MHP