Core stability and low back pain: How stability exercises might help. Part Two

In part one of this post I very simple reviewed some of the ideas behind core stability and how I questioned their relevance to a patient’s pain presentation.  In this follow up post I will briefly review how people with pain have different function than those without pain and give an opinion on how core ‘stability” exercises may help with patients in pain in a manner that has nothing to do with stabilizing the spine.

What do we know and what can we do with patients with low back pain

- some patients with low back pain show delays in Tranny firing

- this delay can be correlated with the presence of low back pain

- changes in muscle timing occur with perturbations to the spine

-improvements in pain and function can occur and have nothing to do with changes in firing onset (see here )

-changes in firing onset can occur and have nothing to do with a motor control retraining plan (see Gary Alison’s recent work here which he has been trumpeting this idea for more than a decade, here and here)

- some clinical prediction rules help identify who best responds to a “spine stability program”

- I published two literature reviews years ago that outline how peoples spines function differently with pain (Here and here). The research shows those with pain have differently behaving muscles, changes in proprioception, differences spine kinematics and differences in how they move

But so what if there are changes in function?  Do we need to know this to make improvements in pain?

None of the research suggests that these differences occurred before the pain nor does any research suggest that changing these dysfunctional parameters is necessary for pain resolution.  We probably don’t need to specifically address these changes in function with our treatment and most of us aren’t.  From the above list there are at least 8 means that the spine is “dysfunctional” as measured with some advanced biomechanical testing.  All of us do not do these testing on our patients yet we are often able to make them feel better.  What does this tell you about advanced testing? It probably isn’t necessary.  Or is the test that is most important just the one that your clinic happens to have access to (I’m talking to you you people with Rehabilitative Ultrasound Imaging - I guess you have to justify the cost somehow :) )  We may just need to address the patient’s pain, their beliefs, their attitudes, their activity levels and their habits and we will get changes in these functional outcome measures or we won’t get changes in those outcome measures and we don’t even need to.

These changes in how muscles work can definitely occur in the presence pain.  Where we are confused is correlation and causation. When did these changes begin in relation to onset of pain?  We have some suggestion that changing these motor control variables does not relate to improvements in symptoms (Mannion 2012).  So are these changes just a defense of the body rather than some defect?  They are assumed to be surrogates for stability but are they?

Maybe we should see them as secondary casualties in how our brain works.  The brain changes with pain.  Pain is an output of the brain.  So is motor control.  We can also see changes in swelling, heat, blood flow.  No one thinks that we have to address these secondary adaptations to help with back pain but we think we have to address the motor control or stability issues.  All of which can be secondary byproducts of the persistent pain experience rather than some original criminal mastermind of the patients pain.


What do I think my exercise prescription is doing

Breaking habits of motion, changing fear and building self efficacy

Directional preference, flexion intolerant, extension intolerant, activating neurotags etc. Patients have movements that hurt.  Sometimes they continually perform these movements and they keep hurting.  Maybe they were told that they were supposed to sit up straight, brace their spine, suck in their belly and always activate their lazy glutes.  And guess what, sometimes you have these patients relax, move their spine, sometimes slouch, sometimes put their feet up, stop worrying about their glutes and keep active and voila their pain feels better. Instead of “correcting” some stability problem we just gave the patient permission to move with a lot of variety.   We just broke a pain habit.

Sometimes, patients get pain with flexion and for some reason they keep doing activities that flex their spines and they keep having pain.  Sometimes, we suggest that they move the opposite direction, try to find a position of relief and every hour they arch backwards for a few weeks.  Then we slowly have them start flexing their back again (because we don’t want them to be afraid of flexion, its what we are meant to do) and now they can flex their spine without pain.  Are these patient’s spines any more or less stable? Did their tranny start firing earlier? Who knows, they have less pain.  They began to move differently than what they were previously doing and this helped.  We broke a habit, found positions of relief, built confidence that they could move without pain and transferred that confidence over to other activities.

In some patients you can give them a full spine stability program aimed at buttressing their entire spine. The exercises feel good when doing them.  The patient gets more confident.  They feel stronger and even their pain decreases.  Did their spine’s stability become more robust? Maybe.  Is this what caused less pain?  Probably not.  You got this patient moving, maybe you caused exercise induced analgesia.  You put them in control and they felt better.  It probably helped if you didn’t tell them there spine was unstable before you gave the exercises.


What is my point?

Keep it simple. Adam Meakins (A sport physiotherapists at  wrote a simple tweet


“In physio you cant go wrong if u do the simple things exceptionally well & save the fancy crap for show offs & bullshitters #ROM #power etc “

This might encapsulate my treatment philosophy in all its foul mouthed glory.  While the body is extremely complex and the pain experience difficult to fully understand our interventions can be quite simple.  We are not car mechanics where we are tightening something or loosening something.  We just provide some input into the body.  The body and brain then decide what to do with it.

I still advise bird dogs, side planks, front planks, curl ups, squats and all the exercises that are in the traditional North American stability paradigm.  Sometimes, I even check to see if you someone can suck in their belly button without other muscles turning on and if doing that decreases the pain in their back when they move their leg.  Do I think that any stability changes occur and this is causing their pain to decrease?  NO. There are other reasons that these movements help.

Do I think that all those planks are creating rigidity in the spine.  Of course not, this is much too simplistic.  So what the hell do I do and what do I think I am doing.

My approach and my rationale

All of the following assumes I have ruled out the nasty stuff.

1. Educate about pain.  Don’t catastrophize.  Explain the difference between tissue injury and pain.  Explain that we are meant to move and that pain is normal and is not some indication that they are falling apart.  Explain that their scary x-rays and MRIs are poorly correlated with their pain.   Explain that pain is so much more than just the tissues in their back and reassure them that they can do something about it.

2. Touch them with your hands:  Move them, push, pull, rub, crack, traction, distraction, compression. Whatever.  Manual therapy has some neurophysiological pain modulator effect.  Skip the bullshit explanation about the complexity of the SI joint having an upslip, downslip, a shear or flare or whatever.  You can’t feel this and you can’t correct it.  But using your hands can modulate the perception of pain and can change muscle and joint performance.  This gets your patients confident that change is possible.

3. Move meaningfully.  What is important to your patients?  Find some movement related goal that is important.  Figure out a way to do this.  Set small goals related to this movement and achieve them.  They might have pain during this task but they know that pain does not mean damage and that they can do it.  Don’t hammer them into fighting through pain so that they feel “Wind Up” the next day.  But keeping pushing that pain threshold up.  Keep empowering them.

4. Stress the body.  This is where I use exercise.  I find a movement that is painful.  I figure out someway to modify that movement so that it is not painful (think Mulligan).  Train that movement.  Start to break the habit of pain.  Pain is a habit.  If we activate a pain neurosignature with certain movements we can sometime modify that movement so that that neurosignature is not activated.  Do this.  This is where “spine stability” exercises can come in.  Get them working their achey back in ways that don’t cause pain.  This is awesome they train a painful area without experiencing pain.  This decreases the threat associated with movement, decreases kinesiophobia and changes how they think about pain.

5. Train harder: pick a movement that is kind of related to their painful site but does not hurt at all.  Train the hell out of this.  This can be build confidence in their body.  For example, they might have shoulder pain but they can deadlift.  What a great shoulder exercise.  You work your entire body, train the shoulder but experience no nociception.  Hammer this.

6. Address beliefs: we need to understand what our patients think about their condition and how that impacts their psychosocial profile.  If they have some serious catastrophizing, fear avoidance, depression, perceptions of injustice etc this shit needs to be addressed.

An opinion on motor patterns

I have seen “faulty” motor patterns and I have also seen them “corrected” by doing exercises that have nothing to do with retraining the supposedly faulty muscles.  If a motor pattern is corrupted this pattern is most likely corrupted at the level of the brain.  If I train some movement, modulate pain with some education or mobilization we often see changes in these motor patterns.  But I didn’t change these as a mechanic.  There was no tweaking at a local level of “muscle imbalances”.  We aren’t bloody puppeteers The imbalances get “corrected” via other means.  Or they don’t get corrected and my patient is pain free and I know that “muscle imbalances” can certainly be normal variations of your complex system.

So just remember this mnemonic: KISDAS.  Keep it simple Dumb AsS.


Future posts and questions for research

Corrective Exercise: this approach, besides being an unethical cash grab for CE dollars by some questionable organizations by making good intentioned personal trainers and physios feel insecure, has a number of assumptions about human function that we all need to question.  I also want to investigate it because it is an approach that I kind of use daily (see my defense here of the lowly clamshell) although I try to simplify it and pain neuroscience it up a bit.  And besides, isn’t “corrective exercise” what every good coach or therapist does automatically? Find a deficit and improve it?  Anyways,  corrective exercise has a number of interesting things that we can look at and I don’t think it is worth completely discounting.  Here are some areas I hope to address:

1. It assumes that there is an ideal way that the body functions.  I think this is everything most physios, trainers and coaches do yet our research is so piss poor.  I would love a catalogue of articles that attempt to categorize the best way to move.  We need to settle this debate once and for all.  The catalogue might just show exceptions and the huge variety that is acceptable.  It might show that their instances when it is best to move on way over another.  I know it is more complicated than just avoiding knee valgus, keeping a neutral spine, not letting ‘global movers” shut down “local movers” and assuming that all asymmetry is evil.

2. It assumes that isolated testing (e.g single leg squat, prone leg extension, single leg glut bridge) gives us some insight into altered functioning.  It then assumes that the altered functioning in some isolated test actually correlates with assumed altered functioning during some more meaningful performance task (e.g running, deadlifting, squatting).  I wrote a PhD proposal on this area and you would be surprised how poor our tests are at actually testing anything we think they are testing and also correlating with functional activities. Guess what? Did you know that the quadruped rock back test coupled with spine rotation actually does not “lock out” the lumbar spine and only result in thoracic rotation?  Crazy.  Or that the thoracic spine has just about the same amount of rotation capacity as the lumbar spine.

3. Muscle activation?  Such a neat, simple and prevalent idea.  Training some small movement (e.g squeezing your glutes during a bridge) while “turn on” some muscle during another activity.  I would love to see this idea put through a simple experiment.  Flippantly, it again views the body as something that is so stupid.  Certainly worthy of some good research.

4.   I would love to see a series of blog posts look at altered joint kinematics (what a corrective post assumes it is correcting) and how these alterations correlate with changes in pain or improvements in function. My big hesitation with the corrective exercise approach is that it complicates things and does not seem to recognize the uncertainty that exists in human function. Further, corrective exercise assumes that there is a limited way to “correct” the dysfunction.  aka. you need the “correct” corrective exercises.  I would suggest that there are a multitude of exercises or approaches that can influence pain perceived to becoming from a joint and the resultant aberrant kinematics.  Anyone want to look at this topic?

5. Of course we could also talk about foam rolling.   It is interesting that there does seem to be a shift in the rolling world to get away from the idea of digging out knots and adhesions and focusing on the possible neural aspect for the treatment mechanism.  But fascia is still king in some circles and I can’t fathom why.

I would guess that if we look at the biomechanics and motor control literature we would find that our treatments can be much simpler, we would have lots of variety and many approaches would be successful.  We would not have to have incredible complicated solutions to simple problems (avoid the Rube Goldberg trap of exercise prescription).


Happy new year!  Anyone interested in collaborating on blog posts please email.


Share this
10 Comments - Leave a comment
  1. rachel_a says:

    Thanks for this post. It’s confirmed some of the ideas I’ve tried to convey to a few chronic pain clients with good success. A case of a recent client who fell and injured his shoulder, and several weeks later still had pain, went to a physio was tested/MRId etc with the answer “there’s nothing wrong with it”. He came to me weeks later still in pain and ROM showed limited movement with pain, but to me it felt more like guarding. I also sensed he was very frustrated and tired and almost resigned to the pain.

    I talked to him about his brain “thinking it still needed to protect his shoulder and creating pain signals that he had no control over” - did lots of gentle movement, got him doing some movements at home between treatments that didn’t cause any pain, and hey presto within 3 treatments he had significantly better PROM and AROM, but most importantly his attitude towards the pain changed. He felt like he could do something about it, rather than feeling he was stuck with it.

    As a relatively new MT I’m excited about how much I can learn from others, how much my limited experience has already taught me, and how my views about aspects of my training have changed in just 3 years.

    I shall continue to read your posts with interest :)

  2. ben says:

    nice post Greg. I’m finishing second year chiro and we recently had to present on O’Sullivan’s LBP classification…tough reading but it really opened my eyes abt that there’s a lot of heterogeneity in chronic LBP and at the end of the day, much we do’t understand. eg, as in O’Sullivan’s research - some patients have pain when going in extension. When EMG was measured while they were asked to sit like how they normally sit, then sit *slumped* both EMG also had high readings. Why? Dunno is the answer.

    Okie, now that you have debunked many myths abt core training, etc how do you train then to ‘correct’ swayback, flatback etc? I have read Kendall’s recom, but in light of all the new research/opinions, I’m not to sure.

    For the local vs global muscles, in my opinion we can simplify into superficial (RA) vs deep (multifidi, TA). If you’re doing old school exercises (Crunches, leg raises, jack knives, etc) you’re probably using superficial muscles. I have seen people with six pack abs who may shiver when doing a single leg lift / squat. hence, I feel that this is where there is benefit of learning how to activate the deep muscles ala Diane Lee’s thinking. Heavy deadlifting you’re probably gonna need to use superficial n deep, valsalva, etc.

    I’m going off-tangent but what are your thoughts abt the use of real-time ultrasound for assessing hip stabilisers and exercise rehab? Some Aussie gurus (Alison Grimaldi, etc) seem to be have good results.

    Great post!

    • Greg Lehman says:

      You got three points in there. Pain, performance and hips.

      Pain is different than performance and may be different than injury prevention. If you have an athlete that can do a movement and you train them in a way that gets them to do that movement that is great. You may have identified a limit in their ability. I am all for “core” training in this manner.

      YOu asked “Okie, now that you have debunked many myths abt core training, etc how do you train then to ‘correct’ swayback, flatback etc? I have read Kendall’s recom, but in light of all the new research/opinions, I’m not to sure”

      I don’t think we correct swayback, flatback by doing something mechanical. Meaning I don’t think we stiffen things and loosen things and then thing “get balanced”. I think you can encourage people to consciously move in a different way. This could “correct” those spine positions you mentioned. In my opinion, I don’t think those positions are terrible, my issue would be one of variety and choices. If a patient keeps choosing to use their spine in the same manner and this keeps being associated with pain then that is something we work on. We help them find some other patterns of movement and this can be associated with less pain. I doubt it is because they are in a “better” posture but rather a different posture. This is about changing habits and building variety capacity.

      As for Dr Grimaldi. I read here paper in ManTher and tried listening to a podcast during running. You mentioned good results. I don’ think this is published. I always get a little concerned with overly complicating things with technology. Its fun to use, when you have it you think you need it but it never really changes much of your treatment. If it does change your treatment you probably can’t prove that the treatment is better. Again, pain is so much more than an atrophied muscle, some degeneration, delays in firing etc.

      I keep it simple because things are complicated enough. I don’t think that even our most complicated assessments can actually match the complexity of the human system. Especially when that system expresses pain.

      Hows that for some wishy washy answers!


  3. Jim Ruetenik says:

    Thank you for this description. I’ve been interested in learning more about the non-mechanical rationale for the development of pain. In this and others discussions, I’ve seen much criticism of the weakness of that studies that validate the mechanical approach, but I haven’t seen many references that validate the non-mechanical approach. For example, how do you know that people repeat painful movements? Clinically and personally, I haven’t found that to be the case. For example, I have trouble extinguishing a limp once the pain has resolved, but I’ve never had to teach a limp to avoid pain. If a person’s back hurts when standing, I can’t remember having to tell them to sit down.
    I am open to learning more, but when I see, what appears to me to be a condescending attack on the traditional mechanical approach, it makes it hard to be open to learning more about the non-mechanical approach. I’m sure that’s not your intent, so I thought I would point out my reaction.
    I think there is a lot of validity to the non-mechanical approach, but often times, after reading an article about it, I’m left without a clinical approach to implementing it. I like that you have gone the next step and said that, ultimately we probably do the same thing, we just present it diferently. That makes sense and gives me an approach in the clinic.
    Thank you for your information. I look forward to learning more.

    • Greg Lehman says:

      Thanks Jim,

      I am disappointed that it came off condescending. I will certainly work on this. I am very mechanically oriented. This is my background and training. I don’t completely deny its utility. This what I am exploring. I don’t have all the answers and I know that no one does. The goal of my writing is to keep the lines of communication open. I am hoping to not come across as a know it all asshole. While the asshole might be appropriate the know-it-all certainly isn’t. When I attack the traditional mechanical approach I attack myself. I challenge my own beliefs and hopefully I get better at knowing what I know.

      And you are right after reading an article that questions the mechanical approach we can be left with a void on what to do with our patients. Simply, I would suggest that you can do much of the same but you can start be explaining it differently. You can also start by understanding pain and explaining more about pain to your patients.

      I will close that it is certainly easier to tell patients that their joints are out of alignment, their spine is unstable, some muscle is not firing and I am going to correct it. I still crack-em, rub-em and train-em.

      Future posts will look at explaining more of what we do and explain other things that we can do.

  4. Adam Rufa says:

    Great post. You summed up exactly how I think in the clinic. We create so many false complexities in an attempt to explain why patients have pain or do/don’t get better. Yet, we struggle to find a particular treatment or treatment approach which is far superior to others. It makes me wonder if our varying interventions mostly work via the same mechanism. Most of the great PTs I have observed have one thing in common and many things in opposition. The commonality is their ability to connect, communicate and gain the trust of their patients. Their treatment approaches are often all over the place yet they all seem to get great results. It makes me wonder if the most important thing is not what we do, but how we interact with our patients.

    • Greg Lehman says:

      Thanks Adam,

      I have heard a little of this before but don’t know the research that well. I think if you look up Jason Silvernail and Diane Jacobs on their writings about Operator versus Interactor models of health care you might see some your ideas echoed. I think the research on the non-specific effects of treatment would also support this idea.


  5. Greg, I really appreciate your blogposts. They always get me to question my own beliefs, and often lend confidence to my own beliefs. If I happen to disagree in any circumstance, I never have research to support it - and it makes me realize that we need future research to shed light on some of these topics. If I was a multi-millionaire I’d assemble a lab and a dream-team of sports scientists and the like, and you’d be invited. Cheers, BC

Leave a comment

Your email address will not be published. Required fields are marked *

nine − = 1

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Welcome , today is Saturday, January 19, 2021