I have sat down to write this blog every week now for about 3 months; is this a serious case of writer’s block or something deeper? This is a particular debate, which looms large in our profession and musculoskeletal medicine in general, and not for one second do I propose that my opinion is weighty or even valuable in this argument, but I wanted to write something that I would be happy to look back on and know that it was written with honesty and clarity as to any pre-existing personal biases. Because in the argument for or against manual therapy, this is something that is not done by either side very well. Pain, and your own treatment of it is an emotive issue, and it is YOUR practice that this impacts, the way YOU treat people every day; of course, YOU are going to take along some inherent bias and whether you like it or not. This is going to impact how you will read the evidence.
In all honesty, my exposure to various groups on social media sites has me thinking I’m brain-washed against manual therapy, so I set about looking at the literature on manual therapy with a view to have an objective stance, and really just look for once at the conclusions that get drawn in the literature. But, I will warn you, this blog is written by a physiotherapist with a bias towards exercise. There, I said it.
And I have gone about unpacking this topic and trying to form a slightly different perspective on the debate; that is, as a sensible mid-career physiotherapist with the aforementioned bias, ‘Is manual therapy extinct-as-we-know-it or here to stay?’ What do we do with it as we lead the profession forward? Do we throw the baby out with the bathwater?
The dichotomising of the issue is difficult to swallow for me and I hoped to write this blog to come to a more moderate conclusion on the issue, perhaps a third way. As always, this is an opinion piece, with a weighing up of current literature; make of it what you will and I encourage you to comment and share if you want to continue a more moderate debate on the topic.
The one caveat that I have to stipulate for this debate is that I have gravitated towards a review of the literature on low back pain (without radiculopathy), as it has become a topic I’m regularly blogging about and one of the most burdensome conditions facing our society; kind of important.
Manual therapy is incredibly varied
Something I have heard more than a few times from other physios and a common thread presented for the use of manual therapy is that it is as heterogeneous as low back pain presentations and any research investigating it cannot approximate the intricate factors that go into successful manual therapy, like experience and confidence.
It is true that manual therapy does tend to get lumped into the same basket; in various clinical practice guidelines (CPG), spinal manipulative therapy (SMT) will be lumped in with mobilisation, see 2012 JOSPT CPG (Ref) and a most recent Danish CPG (Ref). But in others these vastly different approaches are correctly separated – see Australian guidelines by Australian Acute Musculoskeletal Pain Guidelines (Ref).
I can see how this might be a frustrating scene for a manual therapist, but exercise therapy is very much as heterogeneous as manual therapy, so I’m not sure how much this argument stands up. Come to think of it, pretty much all of what a physiotherapist does is so non-specific and convoluted that it may not ever be possible to research what exactly makes us successful, if we are at all.
So, in all of the various types, what type of manual therapy is best?
In acute pain: A very objective and fitting review, the comparative guidelines by the Agency for Healthcare Research and Quality group in the US, found when spinal manipulation was separated out, it seems to come to a higher effect size when compared to sham or no treatment (Ref).
Cleland et al 2009 has shown that manipulative thrust better than non-thrust (compared to your bog-standard lumbar PA) in acute low back pain (less than 16 days n=112). Cruser et al in 2012 designed a fairly low quality trial, in military personnel, which makes a pretty outrageous conclusion that osteopathic manipulative therapy in soldiers with a new onset of low back pain is effective in reducing pain. This is disappointing for me, as when looking at the results, it is pretty clear that the one consistent statistically significant aspect throughout the follow-up points is actually just the interaction of time itself. A study in natural healing at its finest.
Another example of a study which seems to be unable to distinguish between natural healing and treatment effect is by Hancock et al in 2007, this is another study used by the most recent Danish guidelines for the treatment of low back pain to back-up the recommendation for the use of manual therapy in acute low back pain.
Hsieh et al in 2002 couldn’t seem to distinguish between three types of manual therapy when applied to 200 people with subacute low back pain randomly. When their four treatment arms were analysed as a collective, joint manipulation, myofascial therapy and combined joint manipulation and myofascial therapy, no group was found to have had a superior effect on pain or disability with 3 weeks and 6-month follow-up.
Chronic: SMT does seem to have a better effect in a good trial by Goldby et al 2006 in pain only (not in function or even quality of life). This was amongst the best studies I’ve read on manual therapy and is one I will use again.
But a bogus study by Aure et al in 2003, demonstrating superior effects of manual therapy over exercise should serve to demonstrate the confounders involved in all low back pain research; they found significantly better improvements in a small cohort (n=49) in not just pain, but function and disability. It is fairly demonstrative of a typical situation of experienced manual therapists pitted against exercise therapists, the exercise therapists are not allowed to put hands on and the patients with chronic back pain, who have probably had benefit from manual therapy in the past feel robbed, and the manual therapists in the study do their thing and also instruct patients to continue to move more, provide reassurance and reduce fear avoidance.
Physios seem to be good at it
Menke’s awesome 2014 meta-analysis pooled treatment effect sizes of 56 studies involving manual therapy; this involved 6397 measurements of patients with acute low back pain and 2455 measurements of patients with low back pain for more than 12 weeks. In acute pain settings, physiotherapists seem to have the best outcomes when administering spinal manipulative therapy in 22 studies (Ref).
That has to count for something, right? The pessimist in me thinks that physiotherapists are just better at designing robust trials with more validity and, heck, maybe we even give better chocolates to study participants over chiropractors and osteopaths. But seriously, maybe this is another reason why manual therapy is so emotive and hard to separate in terms of fact and fiction; the clinical guru-ism is still at-large!! Clinical treatment is a very special thing and something that people want to be good at, plus it’s kind of good for your bank balance.
Does manual therapy help pain in the first 4-6 weeks?
Going straight to the top of the evidence ladder, let’s have a look at the Cochrane review performed 2012 on acute low back pain (Ref).
‘No high-quality evidence was provided for any comparison, outcome, or time interval; therefore, no strong conclusions or recommendations can be made for the use of SMT for acute low-back pain.’
It is worth noting, that only 8 of 16 included studies were published later than 2000. This major limitation of the study is even accepted by the authors. Does manual therapy from the 1970s bare resemblance to the manual therapy performed currently? Some might say yes, but I would think some things have moved on.
Depending on who produces the clinical guideline and other assorted research, it appears manual therapy may have an effect on acute pain. In Koes et al 2010’s summary of national clinical practice guidelines (CPG) 11 of 15 national CPGs support the use of manipulative therapy in the first month of low back pain (Ref). Is this a sign of a broken system or should this be taken as a sign that manual therapy has some effectiveness?
Small effect sizes and non-specific effects
It’s not for a lack of fair-quality trials that the debate still rages amongst social media and scientific-literature-at-large. The above clearly shows that there is no paucity of evidence for manual therapy; all randomised control trial and all with pretty large samples (>100).
Effects in these studies, however, are generally small changes; generally, a couple of point changes in VAS scales and rarely changes in validated questionnaires. When pooled together, these effect sizes generally get diluted down to insignificant (see Cochrane reviews). Menke’s 2014 review makes this ‘diluting’ effect fairly clear when added together with the myriad of other non-specific effects involved in any treatment scenario. When other factors like placebo, contextual, regression to the mean and natural history all together are accounted for, the treatment effect in manual therapy interventions over the years only amounts to 3% of change.
Wow 3%; most of us wouldn’t scratch ourselves for 3% change in any symptom. Again, it is pretty sobering to think that natural history and simply just therapeutic alliance results in most of the results we see day to day,
‘Results here support cautious observation, monitored exercise, and authoritative encouragement—services not requiring a licensed professional.’
And herein lies the crux of the argument against manual therapy. I continue to ask myself is this 3% clinically relevant when guiding a case down the path toward discharge and empowerment?
Do we intervene early?
Good quote from Childs et al’s 2012 review of the literature (Ref):
‘Current CPGs for LBP mostly recommend delaying referral to physical therapists for at least 4 weeks following initial primary care consultation. This “wait and see” approach is based on the belief that most patients with LBP will recover rapidly, and intervening quickly would not be cost-effective. Furthermore, it is believed by some that early intervention may impede recovery for some patients by excessively “medicalizing” the condition. However, the evidence clearly indicates that this belief and approach to managing LBP must be challenged.’
In my opinion, if done well, LBP is not over medicalised by early referral to health professional. Appropriate education and advice can go a long way to changing beliefs, for the better, for long-term. If done poorly, however, this is exactly what happens, and what has been happening for 30 or more years; conditions do get over medicalised and patients do get dependent on someone helping them out of pain.
How much of this is directly manual therapy’s fault? I think the same problem presents itself with exercise as well; there is still the element of patient interaction and treatment seeking. If you use a sledge hammer or a wrecking ball, you can still demolish a house.
So then, is exercise perfect?
Taken directly from the Cochrane review performed in 2005 on exercise therapy in NSLBP (Ref);
‘Exercise therapy appears to be slightly effective at decreasing pain and improving function in adults with chronic low-back pain, particularly in healthcare populations. In subacute low-back pain there is some evidence that a graded activity program improves absenteeism outcomes, though evidence for other types of exercise is unclear. In acute low-back pain, exercise therapy is as effective as either no treatment or other conservative treatments.’
Part of the driving force behind writing this blog was to explore and present manual therapy research, on a selfish note to re-orient my bias but also to stimulate more thought amongst mid-career or young physiotherapists like myself and for that reason I won’t labour on exercise literature. Effect sizes in various reviews do seem to be significant and something that, obviously, those who share my bias hang their hat on.
But is it all rosy? I would like to point readers to my original blog post on the pathologising of pain; if exercise prescription by allied health professionals was the answer and the cure, then wouldn’t we have seen an impact in the last 10 years in low back pain burden? Assuming that in the last ten years, some form of exercise has been prescribed by even the most devout of manual therapists, then surely there should be some impact on low back burden of disease.. A good study to illustrate this is one performed in 240 people in Sydney, Australia by Ferreira et al (Ref). One on one delivered manual therapy versus one on one delivered motor control retraining (the dreaded TrA and multifidus!) trumped a supervised group exercise class, where 8 participants could sometimes be involved. Now, we know what happened to the hypothesis that participants can’t find their TrA after a bout of low back pain – hog wash – but it was still better than just getting some patients moving for an hour. There may be one specific reason to this; attention. The sanctum of the cubicle and the treatment relationship has a lot to answer for in this study. It may also have a lot to answer for in low back pain research in general. It may be better to get someone to exercise because it has specific physiological effects yes, but you know what, they’ll probably come back.. For exercises for another injury or ache.. Are those of us who share a preponderance for exercise really better than those who don’t share the same enthusiasm? If the system is set up to let people fail, and repeat business is what puts money in your pocket, what is the difference?
So, in conclusion, I have taken 3 months to go around in circles with my thought process. I’m going to break this down simply and in dot points, because if you haven’t stopped reading by now, you are either: a) in need of some dot points or b) on holiday and have way too much time on your hands or c) strange.
Manual therapy cons:
It creates dependenceIt externalises locus of control over disordersIt is passive and reinforces passive coping strategies, which makes this whole cycle of treatment-seeking even worse.The logic for ‘the concept’ of manual therapy is actually quite flawed: if (musculoskeletal) pain is largely a movement disorder, to treat without any form of movement unpacking and retraining is missing the point. Why does poking someone improve a disorder of the nervous system experienced with movement?There is a dearth of evidence, but not much is high quality and most of it can’t distinguish between natural history with small effect sizes generally seen.Personal bias is huge when you are delivering a treatment and due to myriad of different factors like neurophysiological, placebo, context of treatment, unfortunately the patient and the clinician can both be fooled into spending time on low value manual interventions because it actually causes a reduction in pain. Think ‘lead down the garden path’; it can be carried on for too long and used instead of other more valuable interventions.
Manual therapy pros:
Any form of change must first begin with trust. If you are treating patients to change their life, then you must first build trust.Putting a ‘helping hand’ on someone’s shoulder, so to speak, is a great way to build trust. And for some people, they have such low volition that internal change may never happen without first playing a passive game.In today’s health model, physiotherapy’s identity is still embroiled with manual therapy.Patient expectations dictate treatment outcome and unfortunately the law of averages generally plays out in that a lot of patients will have previously been to a physiotherapist, had manual therapy, gained benefit (by this therapy, placebo, natural healing or other contextual factors) and because an external party delivered this, it automatically has more power than anything internally generated, they have latched onto this as a powerful treatment mechanism and anything else will not be as beneficial in their eyes.In order to effect change, you have to first give the therapeutic alliance the best chance at success and meet these expectations. Because after all, in a typical physiotherapy context, the patient already has preconceived ideas of what to expect; they are coming to a physiotherapist not an exercise physiologist or personal trainer.
So, dinosaur or lung fish?
Until patients begin to come to physiotherapy without those expectations then this clinical dance will continue on. Changing these expectations will take many, many years; it will involve the education of other medical professionals like surgeons and GPs and more concerted efforts by our professional bodies to promote what sets physiotherapy apart from other professions.
To change this perception of the public and ensure expectations do not include passive therapies I think boils down to one simple factor and this is explained by the following statement, ‘if the practice next door stops offering a quick fix for patients’ pain, then I will too.’ This is service delivery at its finest, and it reduces physiotherapy to the lowest common denominator; market forces necessitate supply and demand and while there is demand for passive therapies you would be stupid to assume someone out there somewhere is not going to take advantage of that.
So, for negative reasons, that is why manual therapy is sticking around. Because it can be used to pedal profits and physiotherapists have got mouths to feed. I’ve said before that private practice is probably where physiotherapy can expect to see its long-term destination and public outpatient physiotherapy departments days are numbered purely because it is not a necessary public service (in the privatisation of today’s society, heck even electricity isn’t deemed necessary enough for much governmental regulation, so no offense to public practitioners out there); it’s simply good business to promote some form of manual therapy while the public still expects this.
I would like to challenge the issues that are commonly brought up in today’s social media platforms; if repeat business is your aim, then do we really need to blame manual therapy so much? Or are there much the same issues with exercise therapy, even if the effect sizes are larger? Let me know your thoughts on this blog. Thanks for reading.