Painful Exercise In Rehab – Yes, No, Or A Lot To Think About

Whether rehab exercises should be painful or not has become a recent topic of much discussion. As we have started to value the role of (optimal) loading in rehab, and that we can get patients back to moving and loading pretty early in the rehab process, we have also got to appreciate that this is unlikely to happen without experiencing any pain.

If you have any type of understanding about pain then the simple question of “should we use painful exercise?” suddenly might not look so simple!

Rather than a simple yes or no, there are a bunch of questions to be answered, such as how much pain is OK? Who may or may not benefit? And what is the best way to manage the process? Especially if all does not go to plan!


From a data standpoint it would be great if we had have something that we can use to guide us here. Low and behold we do, and its open access.

Smith et al looked at this question in 2017.

Should exercises be painful in the management of chronic musculoskeletal pain? A systematic review and meta-analysis

This systematic review and meta analysis looked at 9 trials in which varying degrees of painful exercise were used. They found that painful exercise did NOT result in statistically worse outcomes across a short, medium or long term follow up. There was a small statistical benefit in the short-term for painful exercises as highlighted by the standardised mean difference of around 0.2 (SMD = effect size for a meta analysis). So from a research standpoint this is positive news.

One thing to note here from a clinical standpoint is, that using painful exercise is no GUARANTEE that it will have a positive effect for your patient. All statistical tests only give us the PROBABILITY of an effect, so the likelihood is that it will, but when we look at the confidence intervals (measure of variability) for the mean SMD for painful exercise, for the majority of the studies we can also see that they cross into a negative effect and in some of the studies quite substantially.

The variable effect of exercise on pain (although over a shorter time frame) was also shown by O Neill et al’s recent look at the acute effects of isometrics *Here*. Although, of course, isometric exercise and painful exercise is not the same thing, the point here is the highlighting of the variable effects from exercise. For patients with higher baseline pain isometrics actually increased their pain post exercise and this is the benefit of looking at individual responders within a study not just the group average. Those with lower pain at baseline seemed to receive more analgesia from isometric exercise.

BUT at the very least I think we can say that the likelihood is that it will not have detrimental effect. So as rehab can often be painful we can be reasonably confident but without seeing painful exercise as some sort of fire and forget panacea and also appreciate the individual nature of pain.

We could debate the mechanisms of HOW painful exercise might work from pain habituation to conditioned pain modulation and perceptual mechanisms but in essence we don’t really know at this point in time and this might be different or in different combinations for different people.


The benefits can be both physical AND psychological.

Firstly, it allows patients to get in MORE exercise. If they are not being limited by pain (or pain contingent if we are being fancy :) then they can get a greater dosing than if they stopped due to it being painful. It is important to add though that we currently don’t know what optimal dosing is for exercise having an effect on pain!

Secondly, it sends the message that hurt does not equal harm. The problem can be painful and still settle down if managed in the right way. This may give people a new insight into their pain and how they can manage it.

Pain self-efficacy may also be built experientially, so using painful exercise as a tool might be an option to help do this.

Clinical Application

So it’s great to have some objective data on this subject, but as we know getting that into clinical application is not always straightforward.

Firstly don’t be afraid of some pain! The likelihood is that it won’t cause a worse outcome. But as we are dealing with thinking, feeling HUMANS, it is OF COURSE not as simple as this, but more on that later.

One of the issues with always looking to be pain free with exercise or continually treating pain, is what does that say ABOUT pain? On one hand we are telling people hurt does not mean harm but our actions might not be saying the same thing. If pain is OK, if it is normal then in the process of getting it better we might have to endure a little bit of it.

But how much is OK is a key question? Studies that have allowed painful exercise have ranged from not increasing baseline symptoms during or post exercise up to 5/10 on the VAS.

We know the limitations of the VAS as a REALLY subjective scale so perhaps we could introduce the idea of pain being tolerable or intolerable rather than a numerical score.

Screen Shot 2019 01 21 At 13.16.00


Now tolerable might be a bit sore but it does not really get in the way. Intolerable might be thought of as getting in the way of our daily activates, maybe having an effect on our sleep and generally making life a bit miserable. This is going to be different for different folk so hence why the concept of tolerable becomes important.

We would also like to know how the pain is settling down. So pain that is not really calming down after the exercise stimulus or keeps on getting progressively worse over a number of days as the exercises are performed is probably not a great idea. We would like to see a nice stimulus response with only slight increases above baseline (tolerable remember : ) that calms down gradually over 24-36 hours, quite similar to delayed onset muscle soreness (DOMS)

So on the issue of DOMS, it is important to identify if the increase in pain is actually the same as the original symptom. Post exercise soreness is quite normal but for those that are unaccustomed to exercise this might not be recognisable especially for a currently painful area. Defining what is pain and what is soreness might be a good sense making exercise for some patients. Although we don’t know that progressive overload is required with exercise in rehab, it probably ia a good idea, so having some sort of tolerable baseline for patients to progress or regress from is a great idea too. This is also might have an effect their self efficacy using exercise too which, of course, is a bonus.

Pain Is Not The Only Issue

Potentially the most problematic area with painful exercise, and one that is really under discussed, may be more about the psychological aspects that are associated with the pain and how they can affect the person. One of the best prognostic factors for recovery in a number of body areas appears to be pain self efficacy, *here* is a recent paper from Chester et al and also from Foster et al *here*  Pain self efficacy, something I have written about before also *here*, is the ability to carry on normal functioning despite of pain. It is important to note that those with higher pain self efficacy also tend to adhere to exercise better.

So someone’s pain self efficacy is going to be a KEY factor in whether they are able to tolerate having pain and being able to carry on functioning which maybe a fundamental skill at the heart of painful exercise especially if it can take 24-36 hrs to settle.

Predicted outcome is another key prognostic factor and if someone believes that increased pain will result in a negative outcome then this could have a negative effect on the actual outcome.

Jack et al *here* found that the number one reason for poor adherence to exercise programs was that people did not want to make the problem worse. So painful exercise coupled with negative beliefs about pain, low pain self efficacy and a poor predicted outcome may not sit well together.

A qualitative piece “Exploring experiences, barriers, and enablers to home- and class-based exercise in rotator cuff tendinopathy” from Sandford et al also found that a fear of making the issue worse was a key factor in reduced adherence to exercise.

What Can We Do?

Firstly you could screen people for these factors if you feel they are required. We have tools such the Pain Self Efficacy Questionnaire  (PSEQ) and the shorter version the PSEQ-2.

One caveat with using questionnaires, in my opinion : ), is it does allow us to get a score to assess confidence and resilience in managing pain but it perhaps does not tell us about things that are specific to our patients. So being able to weave this into a good subjective that teases these more personal parts of the narrative out is a must.

Also ASK about their expected outcome from the treatment and what would it mean to you if it was painful.

“Do you feel like this exercise will work for you?”

“Do you have any concerns about doing this exercise?”

It might be that someone tells you that they have tried exercises before and they did not work or that they feel it might make the issue worse, especially if it hurts.

It might also be important to discuss with ALL patients what their beliefs are around pain. These might be that pain indicates they are doing more damage to their bodies or they will not be able to work because they have pain or implications for future functioning. It can often be what the pain MEANS to the patient that is the real issue rather than just the sensation of the pain itself.

A key factor to remember is regardless of whether painful exercise is positive or no worse than non painful exercise, a negative belief may stop your patient from exercising


One of the most powerful tools we have at our disposal is EDUCATION, I am not always a huge fan of this term as it can imply a teacher pupil relationship. But in order to provide effective education to people we first have to build rapport and then understand what they may need to know more about.

We can educate on a range of things from the effectiveness of exercise for many issues, that painful exercise is often not detrimental, that pain self efficacy is an important prognostic factor and what pain may actually mean and its relationship with physical damage. It may also be worth talking about the process and what to expect, and often that it is a process of trial and error to get the right level and desired response.

This has to start with understanding the patient, their history and belief structure.

Where You Start Is Not Where You Have To Finish

Do we have to jump straight into painful exercise? I don’t think so. It is not REQUIRED but it is also not harmful.

An important point to make is that for someone with very negative beliefs about pain, reduced pain self efficacy and poor beliefs about the outcome, challenging them right off the bat with painful exercise might not always be the best thing to do. As they build trust in you and confidence in their bodies then pushing into pain progressively might sometimes be a better route.

We know that giving people the option to avoid pain can actually maintain pain avoidance, so whilst not advocating avoidance behaviour confronting these things may take a little time, confidence and education, especially if the person has a lot of negative pain and avoidance based beliefs.

Those that persist in doing painful things that have NOT provide a therapeutic effect may benefit from non painful exercise for a while too. If you feel that pushing into pain has not given the desired response when looking at their history then a break or change in dosage could be just the ticket.

Those with higher levels of pain at baseline on average seem to have worse clinical outcomes so this might also lead me to reason that maintaining this level of pain through exercise might not always be a good idea.

There are no real hard and fast rules here only individuals and individual considerations.

What If It All Goes Wrong???

This is where self efficacy may come into its own. Empowering your patients to self manage maybe valuable in the exercise process.

As we hopefully have told the patient BEFORE they embark on an exercise program I can often be trial and error. Have we enabled our patients to adapt their exercise dosage if it goes a bit wrong? *Here* is a short piece on dosage in rehab

So do your patients know what to look out for in terms of symptoms?

Do they know how to adjust the dosage? This could be in exercise frequency, intensity and time, sets and reps (volume).

Are they able to contact you to ask questions? Do they know this?

I think these are all required aspects of pushing into pain.

A patient I saw last week had religiously persisted at pushing into his painful rehab exercises every 48hrs. He believed if he did what he was told he would get better (the trust!). But after 3 months there was not the desired outcome. In this case he was given no indication what to do if this happened, just blind faith on all sides that he would get better.

Key Points

  • Painful exercise DOES NOT produce worse outcomes from the research we have
  • Research does not guarantee your patients response to painful exercise
  • It is not just about the PAIN sensation! What does the pain mean and how does it affect behaviour such as exercise adherence?
  • Pain self efficacy and predicted outcome are important psychological measures and can be discussed and measured
  • Think about education around the exercise and the pain response
  • It does not need to start with painful exercise!
  • Empower your patients to self manage


Articles Of The Week May 17, 2020

It’s a pretty confusing time when we are all looking at possibly re-opening our clinics. What I like about this post is how it is advocating and lobbying for Massage Therapists and their safety.

“An Open Letter about Re-Opening — to Massage Therapy Employers” – Tracy Walton

We would all be better therapists if we could understand more about the patient experience and what they are dealing with. Which is why articles like this are important for us to read. To truly understand the patient perspective with pain.

“My Time At The Mayo Clinic Pain Rehabilitation Center” – Tom Bowen

Unfortunately, the longer someone experiences pain, the less valid standard MSK assessments and treatments become. This is why we need to start looking at things like education, adopting new narratives, and looking at principles over modalities to help chronic pain patients.

“Rehabilitation For The Persistent Pain Patient” – Eric Purves

There has long been a need for quality research around massage therapy as a profession. Here is a great resource of some systematic reviews that show massage therapy has a growing body of evidence supporting its effectiveness in reducing pain and improving health-related quality of life in a variety of health conditions and rehabilitation.

“Systematic Reviews Of Massage Therapy” – Richard Lebert

As we move forward to our clinics opening, there are many recommendations about wearing masks. While this may be important from a safety standpoint, we also have to take into account how this will affect our non-verbal communication and how it can affect our patients.

“The Challenge Of Surgical Masks For Physiotherapists” – Noi Group


Maintaining Mobility During Isolation


Stress exacerbates pain.

This is a simple truth and isn’t exactly an epiphany to most healthcare practitioners. With the current pandemic crisis, this has become more relevant now more than ever. Everywhere we turn there is another announcement or pandemic response, a commentary on self-isolation or social distancing, and someone, somewhere, looking to sell us a magic cure. 

It’s an onslaught.

I’m an EDS patient, without regular mobility, resistance, and acceleration/deceleration training my pain has this nasty habit of becoming maddening. Over the years I have found that “movement snacks” can be an incredibly effective intervention. The following are general examples of novel movement; just about every movement demonstrated is scalable, and can be modified to suit your needs in terms of your own capacity.

*Please note, my form isn’t perfect, nor is it meant to be. I’m not trying to focus on a perfect repetition so much as a movement that feels good for me. 

Hip Girdle

I don’t think I need to state the prevalence of lower back and sacroiliac pain we see in our clinics day to day. Whether you’re working in a relaxing spa environment or with elite athletes in a competitive setting, lower back pain is practically ubiquitous. The effects of daily stress on low back pain are well documented, and regular movement is one of the best ways we can try to manage that. I’ve found hip girdle based exercises incredibly helpful for this.


The following link begins with some femoroacetabular movement and gradually grows from there to greater complexity and difficulty:


Shoulder Girdle

As an RMT I was taught that stretching the pectoral muscles is the panacea to shoulder and neck pain. Pec stretches do it all: Headaches? Pec stretches. Whiplash? Pec stretches. TMJD? Oh, you’d better believe that’s a pec stretch. 

And they can help, but really stretching is only a very small piece of the puzzle.

For my own rehab, overhead range of motion has always been much more helpful, as well as addressing ongoing issues with thoracic outlet syndrome and management of chronic dislocations. End range exploration and competence is key. 

The following link features scapulo-costal movement, glenohumeral movement, resistance, and coordination to tie it all together:


Spine and Trunk

Strengthen your core. 

What does that even mean? Strengthen it to do what? What structures are we targeting? What directions of movement? For what task?

Core strengthening is practically ubiquitous and all but meaningless, it’s also myopic.

I’ve found addressing trunk mobility to be a much more effective approach. There is the inclusion of strength of course, but more importantly more aware of safety and resiliency in movement.  

The following link opens with some basic control and builds up in terms of simple to complex ranges of motion, and mild to moderate resistance.



These videos are intended as demonstrations of movement exploration only, not interventions for existing pathologies. If you’re uncertain about any movement, consult with a professional.



Chadi C Abdallah, and Paul Geha “Chronic Pain and Chronic Stress: Two Sides of the Same Coin?”

Eric Bowman, “Low Back Pain: What Are We Doing Wrong and What Can We Do Better?”



Articles Of The Week May 10, 2020


We are in an unprecedented time, which is about to get more confusing. Rumour is we’ll be allowed to go back to work soon and we aren’t really clear on what the rules surrounding that will be just yet. A lot of things will be left up to your clinical decision making and we have to be able to work with a clear conscience. And that will be entirely up to you.

“Your Governor Is Not A Massage Therapist” – Cal Cates

Whether we are using telehealth or seeing a patient in-person part of our job is to help motivate a patient with regards to their homecare. Here’s some great advice with regard to motivating patients for home workouts.

“How to Motivate Clients for Home Workouts” – Guillermo Munoz

During this social isolation, it can be hard to figure out ways to stay active. Here’s some great advice (with an infographic to share) on various ways we can not only stay active but also boost our mental health during this time.

“Stay Calm, Be Active: Simple Ways To Boost Your Physical Activity During COVID-19” – BJSM

I’m sure we’re all taking some online classes right now (or perhaps putting some on) just to kill the boredom. Here are some great tips around the etiquette of taking online classes…and yes, one of them is putting clothes on!

“The Etiquette of Online Classes” – Zeel

There is many a manual therapy myth out there when it comes to patients dealing with low back pain. Here’s another classic example along with the reasons why we need to stop blaming a specific muscle for back pain.

“Why We Need To Stop Blaming Transverse Abdominus For Back Pain” – Brendan Mouatt


Why Weak Glutes Aren’t A Reliable Predictor Of Low Back Pain

It seems we are constantly trying to find causes for low back pain with our patients.

Quite often we are caught up looking for biomechanical issues (which there often are) and avoiding other things that could influence a person’s pain.

One such instance is the possibility that weak glutes are a contributing factor.

I recall seeing one of my instructors demonstrate this (what I perceived to be) complicated movement test where my patient was lying prone and having them extend their leg. There was a specific pattern where the muscles in the low back down through the glutes and hamstrings were supposed to move.

This would be followed up with the comment: “your glutes aren’t firing”!

While I watched this all I could think was…”there’s no way I’m going to remember what order all of this goes in”!

But, when done with confidence and a really good explanation, it seemed to make sense to me, and the patient.

So, I worked through the low back, glutes, and hamstrings, and the patient felt better after the treatment.

I proceeded to pat myself on the back for a job well done!

However, the next time a patient came in for what I thought was the same issue, I couldn’t remember this specific pattern of movement and what it was supposed to tell me! Why couldn’t I get this!?

Well, in reality, I really didn’t have to and here’s why.

Limitations With Testing

There are a couple of tests used to see if the glutes are a contributing factor to low back pain, so let’s look at both of them.

The prone leg extension test1 is pretty commonly used to test for low back pain and lumbopelvic function (this is the one my instructor was demonstrating to me).

The pattern of movement you’re supposed to see is (if you were testing the right side) right glute max, right hamstring, left lumbar erector spinae, right lumbar erector spinae, left thoracolumbar erector spinae, then the right thoracolumbar erector spinae. A delay in glute max recruitment is supposed to show a dysfunctional pattern of movement.

There is a whole list of other patterns that are supposed to mean other things, but it’s a long list and more than I want to get into for the purposes of this post.

A studyshowed there was no consistent order of activation during the test and glute max was the last muscle to become activate with time ranging from 0.07 to 0.676 seconds in delay.

Now, these tests were done with proper EMG testing, so I have to wonder…how would we ever be able to properly assess this just using touch and vision!?

Another test we were taught was “Trendelenburg”, or “SLS” to assess for glute weakness possibly contributing to back pain. 

With this test, a practitioner stands behind the patient and has them raise one leg so the hip is between 60°-90° of flexion. A positive test shows lateral pelvic tilt on the stance leg, which is supposed to represent glute weakness on that side.

One study 2 compared the differences between those with chronic low back pain vs. a control group using Trendelenberg. It actually showed the number of positive tests was no different between the groups. There was also no difference in glute med strength between those who scored a positive and negative test.

Since there was an equal number of positive and negative tests between both groups it showed this test is unreliable to differentiate between those with chronic low back pain and those without, nor was it reliable for demonstrating a difference in glute med strength.

Another study 3 actually injected a superior gluteal nerve block to see if there were any alterations in different movement variables. They found after the injection there was no difference in contralateral pelvic drop, hip adduction, or hip abduction due to reduced strength of the glutes post-injection.

The researchers found this surprising as an impaired gluteal nerve has always been associated with the SLS test.

So, in other words, lateral pelvic drop is not associated with weak glute med muscles in our patients who are dealing with chronic low back pain. It was even suggested that glute med strength of less than 10% of body weight is required to actually get a positive SLS.

The suggestion was also made that in order to maintain your stance during this test it is because of all muscles attaching to the greater trochanter working together combined with the mechanical force of the iliotibial band. This shows a positive SLS is more likely a global issue rather than a specific spine and pelvis issue.

In light of their findings they showed how this test could not distinguish between who was experiencing low back pain and who wasn’t, nor could it identify those who had weak glutes, and in fact demonstrated the test was probably due to some other issue altogether.

Another study 4 showed the presence of low back pain was a combination of higher BMI, signs of hip abductor problems, significant glute med weakness, gluteal tenderness, and a positive SLS (but the strength testing was admittedly not reliable in the study). However, this study showed something else that is probably far more significant which we will look into next.

It was unclear if glute med muscle weakness was the cause of the low back pain…or just a consequence of it.

Muscle Activation, Weakness, Or Protection?

Quite often when discussing this kind of thing and it’s relation to low back pain, the discussion can revolve around not only issues with the glutes, but also hamstring length and activation patterns between the two.

Inevitably when discussing “weak glutes” it’s due to an S.I. joint “dysfunction”, or this dysfunction is causing the weak glutes. But really, what is dysfunctional with the S.I. joint in this case?

They 5 described S.I joint dysfunction as; low back pain below L5, pain over the posterior aspect of SI joint around PSIS and buttock with or without above the knee leg pain. They used a combination of three SI joint provocation tests (posterior shear, compression, distraction, and sacral thrust) that showed good sensitivity and specificity (although the sensitivity rating was much higher).

When we look at the discussion around which muscles are “activating” with certain movements it is usually centered around the context of the pain-spasm-pain model which suggests pain results in increased muscle activity, in turn creating more pain.4

Or, the pain adaptation model which suggests pain reduces activation of muscles when active as agonists and increases activation of muscles when active as antagonists.They say this will reduce movement and ROM which would, in turn, prevent mechanical pain and further damage to the tissues.

The funny thing is that while both of these models are trying to prove the same thing, they end up contradicting each other quite a bit and neither of these ideas can predict how back pain will affect muscle activation.

There is lots of discussion around both increased and decreased muscle activation due to pain which results in disrupting motor control. However, this systematic review 6 argues that increases in muscle activation is an adaptation that occurs for various reasons, and those changes are mainly to avoid harmful stress placed on already injured structures, and to increase stability around the spine.

In each case, they looked at any changes that were task-dependent, related to an individual problem, and highly variable between each patient. This also shows those changes are functional since they are trying to reduce unpleasant stress and provide stabilization to the spine.

One common thread through most of the research cited is glute weakness in relation to back pain. However, it’s a bit of a chicken or the egg, is the weakness a result of reduced use because of pain, or did the weakness cause the pain? I’d venture to say it’s due to the reduced use.

While this can all sound maybe a little complicated and like word salad, when we look at everything we know about pain, what is this actually telling us?

One thing we know for sure that pain is ALWAYS a protection mechanism. So, if something in our body has a decreased ROM for a period of time in order to prevent stress in a sensitive area, this is just simply a way to protect that area. If those altered movements are also providing extra support to stabilize something, it’s actually doing this to protect the area! With this understanding we have to ask, does it really matter what order a muscle fires in (especially because we’d never be able to accurately assess this with just the human eye)? What we should be concerned with is calming those sensitized areas down, reassuring the patient they aren’t broken, and educating them as to why they’ll be okay. Then get them moving in pain-free full ranges again and help strengthen the area, build up that resilience!



  1. Lehman GJ, Lennon D, Tresidder B, Rayfield B, Poschar M. Muscle recruitment patterns during the prone leg extension. BMC Musculoskeletal Disorders. 2004 Dec 1;5(1):3.
  2. Penney T, Ploughman M, Austin MW, Behm DG, Byrne JM. Determining the Activation of Gluteus Medius and the Validity of the Single Leg Stance Test in Chronic, Nonspecific Low Back Pain. Archives of Physical Medicine & Rehabilitation [Internet]. 2014 Oct [cited 2020 Apr 17];95(10):1969–76. Available from:
  3. Pohl MB, Kendall KD, Patel C, Wiley JP, Emery C, Ferber R. Experimentally reduced hip-abductor muscle strength and frontal-plane biomechanics during walking. Journal of athletic training. 2015 Apr;50(4):385-91.
  4. Cooper NA, Scavo KM, Strickland KJ, Tipayamongkol N, Nicholson JD, Bewyer DC, Sluka KA. Prevalence of gluteus medius weakness in people with chronic low back pain compared to healthy controls. European Spine Journal. 2016 Apr 1;25(4):1258-65.
  5. MassoudArab A, RezaNourbakhsh M, Mohammadifar A. The relationship between hamstring length and gluteal muscle strength in individuals with sacroiliac joint dysfunction. Journal of Manual & Manipulative Therapy. 2011 Feb 1;19(1):5-10.
  6. van Dieën JH, Selen LP, Cholewicki J. Trunk muscle activation in low-back pain patients, an analysis of the literature. Journal of electromyography and kinesiology. 2003 Aug 1;13(4):333-51.




Knee Tilt Mobilizations – Improve Knee Flexion Past 90

This is one of my favorite old school manual therapy techniques I learned while in Fellowship with the University of St. Augustine.

When my current techniques of tibial IR, lateral tibial glide, or other Functional Mobilizations I do in closed chain don’t work to improve knee flexion, I try this knee tilt.

It can be enhanced or made much more comfortable with an EDGE Mobility Band. Try this mobilization and let me know what your results are!