Randomised controlled trial of physiotherapy compared with advice for low back pain
BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.38216.868808.7C (Published 23 September 2004) Cite this as: BMJ 2004;329:708
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
As both a clinician with 10 years practical spinal experience at a
senior level and being involved in acute back pain research I agree with
the paper on the following points. The sample reflects the mixed group we
see in the NHS. Helen Frosts research included a subacute group (6 - 12
weeks), with 23% in the therapy group and the 25.5% in the advice only
group. The majority were chronic 77% of the therapy group and 74.5%of the
advice only group. 90% of patients recover spontaneously within 6 weeks,
thus this research avoided this group but the question evaluated also the
most chronic group. The many subgroups of physiotherapy treatment in the
therapy group along with the mixed experience and training that the large
physiotherapy population is common in the NHS. Alessandro (2004) in a
recent review indicated that even with specific subgroup of chronic
discogenic patients where they are assessed and treated by a direction of
preference used by experienced and trained mckenzie practioners produced a
centralisation effect in 52% of chronic patients distal symptoms.
As a clinical specialist clinical effectiveness is promoted but outcome
measures are difficult for most clinicans to intrepretate. To identify the
sub group in clinical practice reaching a high level of clinical
effectiveness would require a multitude of predictive and outcome measures
to formulate a possible indication of clinical outcome with experience. In
practice mild or moderate back pain is difficult to define as there are
differing components of their problem i.e. psychosocial, disability or
pain. I consider it positive that despite variable levels of professional
experience and type of back pain that there was patient perceived benefits
for either intervention on the NHS. This paper will encourage through its
debate professionals to review their current practice.
Alessandro A, The centralization phenomenon of spinal symptoms—a
systematic review. Manual Therapy, 2004 Aug, Vol:9(3), 134-43.
Competing interests:
None declared
Competing interests: No competing interests
Another paper has appeared whose title gives an unequivocally
negative message about physiotherapy for back pain. It omits mention of
sub-groups where physiotherapy has shown to be helpful. e.g. twice as many
patients with back and leg pain are , other things being favourable, back
at work the week after manipulation than are controls.
John Mathews
ref:Mathews J. A. et al. British Journal of Rheumatology,1987. 26, 416 -
423
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir,
It is a shame that another low back pain study has failed to address
a primary recommendation in the field of low back pain research, namely
the problem of washed out effects due to the heterogeneity of low back
pain(1). Again, we are left with decidedly under whelming results.
Patients with low back pain and low back and leg pain are a hugely
heterogeneous group. The biopsychosocial characteristics of patients with
non-specific low back pain (NSLBP) are so disparate that there has been
the strongest recommendation to address the issue of the sub-
classification of low back pain(1). It is well recognised that in clinical
trials with heterogeneous samples treatment effect sizes can be reduced as
the large effects evident in some patients are countered by the small
effects in others. By establishing homogenous sub-groups of patients
within NSLBP it has been suggested that this “wash-out” effect will be
reduced and more effective treatments for certain “types” of NSLBP
established(2).
This study made no attempt to address the issue of heterogeneity
within the study sample and compounded the wash out effect by providing an
extremely heterogeneous intervention. As a result we can only be left to
conclude that in NSLBP patients in general, spending an additional hour
and a half in the company of a physiotherapist leads to no additional
improvement in disability 12 months after the cessation of contact. I
don’t think many of us would be that surprised at that outcome.
There is the most urgent need for physiotherapists to accept that
treatment effect sizes in this field will continue to appear small unless
trial design includes the sub-grouping of NSLBP into valid sub-categories.
Until these sub-categories are established we will continue to see under
whelming findings reported in the literature. Whilst this study has shown
some benefit for these two interventions, with subjective and short-term
improvements, it is only a matter of time before the resources needed to
produce such small effects will be strongly questioned.
Physiotherapists spend many years developing specialised skills in
the assessment of patients with NSLBP. Expert clinicians, such as members
of the Manipulation Association of Chartered Physiotherapists have been
shown to identify patterns of presentation quickly(3;4) and accurately and
consider this process of clinical reasoning to be vital. Let us recognise
that this process of sub-categorisation occurs and develop valid sub-
categories within NSLBP before more under whelming research is reported.
Reference List
(1) Borkan J, Koes B, Reis S, Cherkin D. A report from the second
international forum for primary care research on low back pain. Spine
1998; 23(18):1992-1996.
(2) McCarthy CJ, Arnall FA, Strimpakos N, Freemont AJ, Oldham JA. The bio
-psycho-social classification of non-specific low back pain: A systematic
review. Physical Therapy Reviews 2004; In Press.
(3) King CA, Bithell C. Expertise in diagnostic reasoning: a comparative
study. British Journal of Therapy and Rehabilitation 1998; 5(2):78-87.
(4) Doody C, McAteer M. Clinical reasoning of expert and novice
physiotherapists in an outpatient setting. Physiotherapy 2002; 88(5):258-
268.
Competing interests:
None declared
Competing interests: No competing interests
EDITOR- One of the main causes of chronic idiopathic back pain
appears to be magnesium deficiency which would not necessarily respond
quickly to either general advice or physiotherapy. Deficiency of this
essential bulk mineral can be related to impaired muscle function by the
use of a myothermogram. This test, first described by Dr John McLaren
Howard in 1989, uses sensitive temperature recording equipment in a
clinical test of muscle action.1
Dr Howard wrote that in the subcellular events responsible for muscle
contraction the biochemical energy utilized must equal the energy output
of the system and this is essentially the external work done plus the heat
energy produced. The external work can be limited so that a plot of the
heat produced during contraction and relaxation should reflect the
subcellular chemistry. The limiting factors are the difficulty of
detecting very small temperature changes and the efficient way in which
muscle heat is conducted away by the circulation.
The myothermogram demonstrates abnormalities of potential diagnostic
significance in deficiencies of magnesium, calcium, iron (in
children),manganese and folate. The test also detects reduced oxygenation
or
perfusion and abnormal results are seen in thyroid disorders. It shows
that muscle damage can result from exercise during magnesium
deficiency,which can also be diagnosed by analyses of magnesium levels in
sweat and
red blood cells. It is easy to imagine how mechanical back problems can
occur when muscle are contracting irregularly, as demonstrated by
irregular patterns on a myothermogram.
In some patients magnesium may be difficult to replete, especially if
the patient has absorption problems, hypochlorhydria or selenium
deficiency. However, verified successful repletion of magnesium is usually
accompanied by resolution of idiopathic back pain in my experience and
that of others.2
1.Howard J. Muscle Action, Trace Elements and Related Nutrients: The
Myothermogram. In: Chazot G, Abdulla M, Arnaud P, eds. Current Trends in
Trace Element Research: Proceedings of International Symposium on Trace
Elements. Paris, 1987, Smith-Gordon, London, 1989, pp79-85
2. Vormann J, Worlitschek M, Goedecke T, SilverB. Supplementation
with alkaline minerals reduces symptoms in patients with chronic low back
pain.
J Trace Elem Med Biol. 2001;15(2-3):179-83.
Competing interests:
None declared
Competing interests: No competing interests
Sorry for the physiotherapists` public image, but this study in
question is perfectly in line with a massive empirical experience that we
millions of health-care professionals throughout the world experience
every day, while interwieving our new patients. 99% the same story to
tell; "physiotherapy was a waste of time". And the few ones, who tell
something positive about physiotherapy, probably would have got better
just by waiting anyway. Feeling better has so much to do with just time
passing on( body`s automatic healing ) and with the feeling that somebody
is taking care of me as a patient as well.
Here in Finland, as in most other western countries, the situation is
based upon general practitioners´ prescription "order" to receive
physiotherapy for patients`low back pain. Physiotherapists don`t have the
right nor the adequate knowledge to diagnose the cause of low back pain.
But do general practitioners have this highly specialised skill to
diagnose the complex entity of back pain? - To my opinion, no.
How about more educated doctors, e.g. orthopaedic surgeons etc.?- The
answer, unfortunately, in the vast majority of cases is still no. This is
supported also by numerous medical doctors of whom I know personally, and
who are my clients as back pain sufferers.
Chiropractors who have a university education for their profession are
most highly skilled in diagnosing low-back pain conditions. Their
treatment for low-back pain has been shown to be cost-effective in
numerous clinical studies. Patient satisfaction scores have beaten the
competitors every time as well. We know though, that there are always
patients that don`t get better. Usually then the problem is based more or
less on a major structural-functional anomaly or degeneration, biochemical
imbalance of the body or both. some of these conditions we can alleviate,
some not.
We know, that most low back pain episodes fade away if not within a
few days or weeks, at least within 2-3 months, with no treatment. Some
cases tend to remain chronic, though. But is there any sense from tax-
payers` view-point to treat pain sufferes with a treatment modality that
has not securely been proven cost-effective by any standards?- If one
calculates how much time and money has to be spent for a physiotherapy
treatment course of e.g. 15 visits, the sum is amazingly high. The psycho-
social factors behind back-pain are important, but to my experince, count
only for a relatively small group of patients, as a part of a more complex
entity of causative agents.
What has come more evident with my personal experience of round 12
thousand patients until today, is that there more often are multiple
factors behind patients pain. We probably still don`t know all the
contributing factors for back pain, but there has been some progress
during recent years in understanding this remarkably disabling health-
problem. What really is interesting is the realtionship between back pain
and psycho-social behaviour, from a biochemical view-point.
Numerous recent studies have shown that nutrition plays a vital role
in many pain-related conditions. Of course we can not forget all the other
possible causes of pain, like genetic predisposition, structural-
functional anomalies( of which the disc degeneration is by far the most
common cause of inflammatory and acute structural low-back pain ), stress,
ergonomy, exercise, the amount of sleep, traumas, repetitive injury etc.
What do you think, does the patient have any chance of getting a quick
pain relief if his/her inflammatory lab values are exceptionally high?-
Does adipose tissue make us more prone to "silent inflammation" in the
whole body?-Including the brain.
As researchers say today, we may have a situation where our brain is "on
fire", as well as the rest of the body, if we eat incorrectly. That is to
say we tend to get psychological problems as well as pain conditions more
and more in conjunction with each other, if our functional biochemical
state is out of line.
The use of too much carbohydrates in general is the bad thing. Especially
the ones which do have a high glycaemic index(GI). When we eat this food-
stuff in excessive amounts, we get a rapid rise in blood sugar level, and
a couple of hours later a rapid decrease. Blood sugar going up and down on
a regular basis, builds up adipose tissue, which has shown to be a
secretory organ, instead of a passive energy store. Fat tissue secretes
pro-inflammatory cytokines, namely IL-1, IL-6 and TNF-alfa. This leads to
an increased risk of pain conditions anywhere in the body. Logically the
inflammatory hot-spot tends to make it`s nest to the weakest or most
stressed points in the body. Even slightly degenerated intervertebral disc
is a good example of this kind of weak point. If, in addition to
degeneration, the patient with low back pain is not exercising or moving
in any way, the biochemical state of the intervertebral disc will become
more prone to inflammation. This all is true also for slim patients, not
surprisingly.
Some patients may be suffering from low-back pain of severe intensity, for weeks or months. Many of them may become depressed, angry and hopeless because they can`t do anything. So they eat sweet biscuits, because "I thought I have to eat at least something; I do not want to eat anything else, don`feel hungry". In fact they are doing harm with this mixture of sugar and wheat/chocolate- biscuit and not eating properly, both to their brain-chemistry( giving rise to depression ) and to their low-back inflammation. In this kind of situation physiotherapy does not help other than what it may give psychologically some support to patients. These kinds of patients usually will be cured most rapidly by a combination of chiropractic manipulation for the verterbrae that may be completely blocked in the low-back to improve lymph and blood circulation in the area, and a complete change in the diet. Patients should stop eating biscuits at once and start eating a balanced Zone-type diet, plus taking high doses of Omega-3-fish-oil-capsules to fight the inflammation off. Patients should get advice to start walking more and more as the condition gets better. In one week the patient may be back to work. Instead of doing this, if these patients would start a course of physiotherapy, they would be still having their low-back-pain for weeks if not months to come. There is no doubt about it. I have seen this thousands of times during my career. These kinds of positive therapeutic examples of other than physiotherapy treatment are numerous, but still in the mainstream health- care are mostly overlooked and sometimes even ridiculed. We still have, unfortunately, a big problem with how the health-care-sector works in the first place. Medical industry wants their part as well, being by far the strongest financial leader in the health-market. That is something we have to think about: what is also financially reasonable and what is not?- Do we have too much money to throw it out of the window, or are we missing something?
Physiotherapy is, to my opinion at it`s best in rehabilitating trauma
-patients, e.g. after a car-accident or some other trauma. Rehab for
handicapped people as well is a superb treatment of choice in many cases.
Working with athletes is well worth doing, as they need constant care in
trying to manage their career in the jungle of injuries. But in treating
acute or chronic low-back pain( or other musculo-skeletal pain ),
unfortunately it is completely useless.
Competing interests:
None declared
Competing interests: No competing interests
The debate and prejeudices expressed in the various rapid responses
are almost as entertaining as the article itself.
I am not an academic so find it hard to comment on the strengths and
weaknesses of the study. However I did see Ceefax the night it hit the
press and the message stood out loud and clear. "Physiotherapy does not
cure back pain"
No detailed explanation of the pros and cons of the study no background
analysis, of what was done to who or the history of the population
studied. In fact the point that this study explored one physiotherapy
intervention being compared to another would have passed me by had I not
read the full article in BMJ.
We know to our cost in healthcare that shorthand version of medical
studies can create all sorts of problems once the popular media pick up on
the story.(See MMR, etc etc)
Wheter or not this particular study will turn out to have been beneficial
for the long term care of those with back problems, time will tell. But
at least for once physiotherapy is at the heart of the debate and the
profession will be stronger I think for the experience.
Competing interests:
Manager of a variety of Allied Health and Rehab services including community physiotherapy (Which everyone seems to want a piece of)
Competing interests: No competing interests
In response to Dr. Gumaer's comments:
"Chiropractic physicians have significantly better
training in the neuromusculoskeletal arena"
- I think it fair to expect all musculo-skeletal therapists should know
their specific field in more detail than a general physician.
"There is much evidence in the medical literature that
supports the effectiveness of manipulation (chiropractic adjustment) for lower
back pain."
- Manipulation has been practiced for at least 2000 years, chiropractic for
109 years, therefore chiropractic uses manipulation but manipulation is not just
carried out by chiropractors.
"..........and refer to physical therapists for further
ineffective therapy.................... So, please do yourselves a favor,
and refer to chiropractors."
- I was under the impression that chiropractors were physical
therapists, alongside osteopaths and physiotherapists.
A good physical therapist (regardless of the "flavour" of
training) should be able to help these patients regain functional activity
sooner than those who do not receive active treatment.
Competing interests:
None declared
Competing interests: No competing interests
The trial compares conventional physiotherapy with a single session
of advice. It is a trial of two different physiotherapy philosophies
rather than a controlled trial.
The routine physiotherapy group received only a median number of five
sessions, with 82% having six or fewer. The process of "Change of
Behaviour" in patients with back pain may take many sessions. A few
sessions using conventional techniques of physiotherapy may not change the
attitudes of the patients towards promoting exercise.
There is a possibility that this study may convince doctors and
patients and even journalists that physiotherapy is not beneficial. Surely
'advice' given by a physiotherapist is physiotherapy in the same way that
'advice' given by a doctor is still medicine?
The accompanying editorial, in particular the headline and the
wording of the summary are misleading.
Let us not throw away the baby with the bath water!
Competing interests:
AS does not want to lose his physiotherapy service.
Competing interests: No competing interests
Dear Sir/Madam,
Firstly may I take this opportunity to commend the work undertaken by
Frost et al in producing and publishing this research? In this era of
shrinking funding for health care services, increasing accountability to
all the stakeholders in the health care profession and hence the need to
justify the services provided, I believe Frost et al have undertaken the
first steps in answering some key questions. As Albert Einstein said “The
most important thing is not to stop questioning”, and this study has
questioned routine treatment program commonly undertaken in many
rehabilitation clinics/private practices for low back pain. The authors
should also be congratulated for the rigour in the methodology of their
research, as proven when critically appraised by PEDro
(http://www.pedro.fhs.usyd.edu.au/CEBP/), a critical appraisal tool for
Randomised Controlled Trial, their research and its subsequent reporting
scoring reasonably well. Nevertheless, I would like to address some key
issues as highlighted by this research.
My concerns are:
(1). Misleading title: The title of the manuscript indicates
“Randomised controlled trial of physiotherapy compared with advice for low
back pain”. This seems to indicate that the research compared routine
physiotherapy care with advice alone. However, in the very next sentence,
the authors have highlighted that the study compares “routine
physiotherapy” with “advice from a physiotherapist”. Do the authors
believe that advice is not part of routine physiotherapy?
(2). Information on intervention: This research again has highlighted
a commonly encountered problem with most physiotherapy research
publications, lack of sufficient information on the intervention programs.
The manuscript states that “patients in the advice only group……general
advice to remain active” and “patients in the therapy group……a treatment
strategy based on their findings”. As a clinician, how am I meant to
translate this scanty information to my clinical practice, when all the
information I am given is just that the treatment found most effective was
“general advice to remain active”?
Do I, as a clinician, have enough information to confidently, translate
information about the treatment which was ineffective (manipulation,
mobilisation, soft tissue techniques and spinal mobility and strengthening
exercises) and treatment which was effective (general advise). The answer
is NO!
(3). Package of care: Physiotherapists, especially in the treatment
of musculoskeletal conditions, often undertake a multitude of treatments
in one occasion of service. While the composition of the treatment might
vary, depending on the subjective and objective findings on each occasion,
the treatment will encompass varying treatment techniques within one
occasion. This is common Physiotherapy practice and by prescribing a
specific regime, the nature of physiotherapy care is varied. Furthermore,
I note that, the Therapy group also were provided with “advice”. How was
this “advice” different, if at all, to the “advise only” group? Again we
cannot come to a conclusion primarily due to lack of adequate information.
(4). Insufficient numbers: The researchers note that “30% of the
patients failed to provide data for the main outcome at 12 months”. The
numbers of non-respondents at 12 months is excessive (statistically it is
imperative to have >85% as opposed to 70%) in order to come to an
scientifically defensible conclusion. Given that low back pain has a high
reoccurrence rate, this lack of information opens sufficient doubts
regarding the credibility of the results for the long term.
I once again quote Albert Einstein when he said “Insanity: doing the
same thing over and over again and expecting different results”. The most
difficult step in improving practice is identifying what is best practice.
This requires soul searching and questioning theories and practices taught
over decades in many schools across the world. These researchers have
attempted to do that and hence need to be commended for these first, but
important steps. While their research has limitations (which research
doesn't?), its results does support many current thought process within
the Physiotherapy professions and adds strength to the argument of
education and empowerment of the patients in their management of their
problem. While these initial steps are arduous and laden with criticisms,
I urge the researchers to continue and further improve their work in
producing thought provoking and constructive research.
Competing interests:
None declared
Competing interests: No competing interests
Finer measures are required to determine the efficacy of physiotherapy
In my opinion the Frost el al (2004) paper is not a valid indication
of the efficacy of physiotherapy for back pain in general for the
following reasons:
- 26 patients from the advice group were given extra sessions, in
some cases because the physiotherapists felt it was unethical to withhold
treatment(Frost el al 2004). If these people had been given one session
of advice only, they may have altered the statistics
- 74% of the sample had had back pain for more than a year (Robertson
2004). Psychosocial interventions are more important for chronic samples
like this.
- The median number of sessions in the physiotherapy group was 5
(Frost el al 2004). Perhaps this is not realisitic for people who have
had back pain for over a year.
- The Oswestry disability index is a subjective measure of function
relative to pain (Fairbank and Pynsent 2000). The influence of
psychosocial interventions is therefore likely to appear far greater when
this measure is used. It is a valid measure of function but not of
impairment. The absence of pain does not indicate a person has fully
recovered from an injury, and similarly, persistance of pain does not
indicate total absence of recovery.
- The research methodology was not sensitive to speed of recovery
within the first two months of treatment (Duncan 2004).
In short, the study carried out by Frost et al (2004) only suggests
that chronic low back pain patients that physiotherapists are prepared to
advise but leave otherwise untreated are likely to show recovery if
checked at a two-month period, similar to those who have received physical
treatments. I do not believe any other conclusions are valid. The
national press should therefore be ignored.
References
Duncan R.D.D. (2004) What about the early benefits of physiotherapy?
Rapid Responses http://bmj.bmjjournals.com/cgi/eletters/329/7468/708#75668
Fairbank JC, Pynsent PB. (2000) The Oswestry Disability Index. Spine
25:2940-2953
Frost H., Lamb S.E., Doll H.A., Carver P.T., Stewart-Brown S. (2004)
Randomised controlled trial of physiotherapy compared with advice for low
back pain BMJ; 329: 708-0
Robertson G. (2004) Important information missing from the published
article. Rapid Responses
http://bmj.bmjjournals.com/cgi/eletters/329/7468/708#75668
Competing interests:
None declared
Competing interests: No competing interests