Physical Therapy Didn't Work for My Back: What to Try Next

Creator: Low Back Pain Coach
Published: 2026-10-05

Updated October 5, 2026 · Last reviewed October 5, 2026 · 6 min read · 5 sources

Short answer

Exercise helps chronic low back pain by a modest amount on average. In a 2021 Cochrane review of 249 trials, exercise reduced pain by about 15 points on a 100-point scale compared with no treatment, usual care or placebo, at the earliest follow-up (moderate-certainty evidence), and daily function by about 7 points. A course of physical therapy that left your back mostly unchanged does not mean you or your therapist did something wrong. The trials report an average, not a result for every person. In a 2005 meta-regression, programs set for the person, and programs that were individually supervised, each reduced pain more than home exercises alone. ACP lists multidisciplinary rehabilitation and mindfulness-based stress reduction with exercise. WHO's 2023 guideline for chronic primary low back pain lists education, spinal manipulative therapy, massage, cognitive behavioural therapy and NSAIDs with exercise. NICE says to consider manual therapy only as part of a package that includes exercise. The next step is a re-evaluation by someone who has examined you.

Written for adults with long-running low back pain who are cleared to exercise. I have no medical training; this is the method I used and what the research says about it. Some symptoms need emergency care or a clinician before any exercise: check the red flags.

If a course of physical therapy did not change your back, the useful question is what to change next. This page covers what the trials say about why exercise programs get different results, what the major guidelines list besides an exercise sheet, and what to ask for at a re-evaluation. It cannot tell you why your back hurts. Someone who has examined you can.

I had low back pain for about ten years and have no medical or physical-therapy credentials. Physical therapy helped me more than anything else I tried. Every figure below comes from a source listed at the end of the page.

What "didn't work" can mean

Exercise is a modest treatment for chronic low back pain. The 2021 Cochrane review by Hayden and colleagues pooled 249 trials. Compared with no treatment, usual care or placebo, exercise reduced pain by about 15 points on a 100-point scale at the first follow-up (moderate-certainty evidence). The change in day-to-day function was smaller, about 7 points. Those figures are averages across trials. The 95% confidence interval for the pain result is about 12 to 18 points, and the review does not say how many people had little or no change. A course that left you partly better, or not clearly better, does not show that you did the exercises wrong or that your therapist missed something.

What separated programs in the trials

"Exercise therapy" covers many kinds of program. In a 2005 meta-regression of 43 trials in chronic low back pain, Hayden and colleagues compared program features. Against home exercises alone, individually designed programs improved pain by about 5 points more on a 100-point scale, and individually supervised programs by about 6 points more. Programs built around stretching or strengthening did best, and programs that added other conservative care did better than those that did not. A higher dose (more hours of intervention time) fared slightly better, but that gap was small and uncertain. The authors rated the underlying studies as low quality and said strategies to encourage adherence should be used.

Put those features next to your own course:

Was the program set for you, or a standard sheet?
Did someone watch you do it and adjust it, or was most of it home exercise?
Did it include stretching or strengthening?
Roughly how many hours did the whole course add up to?
Could you keep up the home part, and if not, what got in the way?
Did the difficulty increase as the weeks went on? The meta-regression did not test this, and it is the progression question.

If most of your answers point to a short, standard, home-based sheet, that is a different situation from a supervised program set for you that did not help. The why PT exercises stop helping page covers the progression side in detail.

What the guidelines list besides an exercise sheet

The American College of Physicians 2017 guideline says people with chronic low back pain should start with treatments that are not drugs. It lists exercise, multidisciplinary rehabilitation, acupuncture and mindfulness-based stress reduction (moderate-quality evidence), and tai chi, yoga, motor control exercise, progressive relaxation, EMG biofeedback, low-level laser therapy, operant therapy, cognitive behavioral therapy and spinal manipulation (low-quality evidence). If those do not give an adequate response, it says to consider an anti-inflammatory (NSAID) first, then tramadol or duloxetine, and to consider opioids only after those have failed, only if the benefits outweigh the risks, and after a discussion of known risks and realistic benefits.

WHO's December 2023 guideline on chronic primary low back pain recommends education that supports self-care, exercise programs, spinal manipulative therapy and massage, cognitive behavioural therapy, and NSAIDs, as care tailored to the person and often combined. NICE NG59 (updated July 2026) says to consider manual therapy only as part of a treatment package that includes exercise, and not to offer belts or corsets, traction, acupuncture, TENS or ultrasound. The guidelines disagree in places: ACP lists acupuncture and NICE says not to offer it.

The Cochrane review also compared exercise with other treatments. In its subgroup analysis, exercise was probably more effective for pain than education alone (about 12 points) or than non-exercise physical therapy such as electrotherapy (about 10 points). Against manual therapy it did about the same (a difference of about 1 point, with a range that spans zero). So on average, swapping exercise for hands-on treatment was not shown to do better. The chronic back pain treatments page covers manipulation, acupuncture, injections and surgery.

When pain is more than the muscles

ACP and WHO both include psychological approaches: cognitive behavioral therapy in both, and mindfulness-based stress reduction in ACP. Fear of movement, poor sleep and stress are each linked to persistent pain in the research summarized on the fear-avoidance, sleep and stress pages. A program built only around muscles does not address those, which is one reason to ask about a different kind of care and not only a different exercise.

What to ask at a re-evaluation

What do you think is limiting me now, and what in the exam points to that?
Which part of my program was meant to change over time, and how?
What can we measure (minutes sitting, walking distance, reps) so we can both see change?
If that measure has not moved by a date we agree on, what is the next step?
Would a different kind of care, such as a pain-focused program or a second opinion, suit me?

The appointment questions page has a printable list. Asking for a second opinion is reasonable. The trials do not test switching therapists, so there is no number to quote either way.

When it is not a program problem

Go to emergency care now for new numbness, tingling or change in feeling in the saddle area (around the groin, inner thighs, genitals or anus, or when you wipe), new trouble getting or keeping an erection, not being able to orgasm, or a change in feeling during sex, new bladder or bowel changes (trouble starting or stopping pee, a weak stream, not feeling when you need to go, or leaking pee or poo), new pain, tingling, numbness or weakness in both legs, leg or foot weakness that is getting worse (tripping, a foot that slaps down), or back pain that started after a recent serious accident, like a car crash or a bad fall. See a clinician today (urgent care if you cannot get an appointment today) for other new weakness or numbness in one leg or foot, fever or chills with the back pain, new pain after a recent smaller fall or knock, unexplained weight loss or a history of cancer, pain that is rapidly getting worse, or pain that is constant at rest and worse at night, not tied to movement. The red-flag checklist has the full list. A new pattern that does not match your usual back pain is worth a visit even if physical therapy is finished.

What I did after physical therapy

I had low back pain for about ten years, from my early twenties into my thirties. Physical therapy helped me more than anything else, but the exercises came on a printed sheet and never got harder, and when it ended I had no clear next step. I later followed a slow back-extension progression on a 45-degree bench. I started with a few bodyweight reps and over about eight months worked up to loaded sets, doing a little and checking my back the next morning before doing more. My pain went down a lot. Read the story.

Everyone's back is different, so results vary and aren't guaranteed. A Cochrane review of 249 trials found that exercise programs for chronic low back pain reduced pain on average compared with no treatment, usual care or placebo, by about 15 points on a 100-point scale at the first follow-up (closest to 3 months), and improved day-to-day function by a smaller amount. This exact progression has not been tested in a trial. The app is not medical treatment; talk to your doctor before starting a new exercise program, and stop and get help if you have any of the warning signs listed in the app.

Low Back Pain Coach is an app that runs a bench progression with a next-morning pain log. It needs the bench, and the app page has what it does and costs. It does not replace a re-evaluation.

Common questions

Why did physical therapy not work for my back pain?

There is no single reason. Exercise helps chronic low back pain by a modest amount on average, so a partial result is within what the trials show. The program may not have been set for you or supervised, the difficulty may not have progressed, the dose may have been low, or something other than muscle strength may be driving the pain. A re-evaluation by a clinician who has examined you can sort out which.

Should I try a different physical therapist?

A second opinion is a reasonable step. The trials do not test switching therapists, so no number says how often it helps. A different clinician may examine different things, set a different dose or suggest a different kind of care. Bring your old exercise sheet and your notes on what changed.

Does exercise work for chronic back pain at all?

On average, yes, by a modest amount. A 2021 Cochrane review of 249 trials found exercise probably reduces chronic low back pain by about 15 points on a 100-point scale compared with no treatment, usual care or placebo, with a smaller effect on daily function.

What do the guidelines recommend if exercise did not help?

ACP 2017 says to consider an anti-inflammatory (NSAID) first for chronic low back pain that did not respond to treatments that are not drugs, then tramadol or duloxetine, and opioids only after those have failed. Its non-drug list also includes multidisciplinary rehabilitation, mindfulness-based stress reduction, cognitive behavioral therapy and spinal manipulation. NICE says not to offer opioids for chronic low back pain. Talk to a clinician before changing any medicine.

When should I see a clinician instead of repeating my exercises?

Sources

Hayden JA, Ellis J, Ogilvie R, Malmivaara A, van Tulder MW. Exercise therapy for chronic low back pain. Cochrane Database Syst Rev. 2021;9:CD009790. link
Hayden J, van Tulder MW, Tomlinson G. Systematic review: strategies for using exercise therapy to improve outcomes in chronic low back pain. Ann Intern Med. 2005;142(9):776–785. link
Qaseem A, Wilt TJ, McLean RM, Forciea MA. Noninvasive treatments for acute, subacute, and chronic low back pain: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2017;166(7):514–530. link
World Health Organization. WHO releases guidelines on chronic low back pain. News release, 7 December 2023. link
National Institute for Health and Care Excellence. Low back pain and sciatica in over 16s: assessment and management. NICE guideline NG59. Published 2016, last updated 29 July 2026. link

The app

Low Back Pain Coach runs a guided strength program on a 45-degree back-extension bench: floor work until you have a bench, then four stages from short holds to weighted reps. You log your pain the next morning, and that log sets the next session. A voice counts the holds, reps and rest.

Your first guided session is free. Then $59.99 a year or $9.99 a month, renewing until you cancel, with a 7-day free trial for eligible new subscribers. US prices. iPhone and Android. The 45-degree bench is required. What the app does.

Who wrote this. The person who built Low Back Pain Coach, after about ten years of low back pain. No medical or physical-therapy credentials: health claims link to trials, reviews and clinical guidelines in the Sources list, and corrections go on the corrections page. Last reviewed October 5, 2026. How these guides are written. Published by Scrolling Is Evil LLC.