Rapid Responses to:

RESEARCH: K J Thomas, H MacPherson, L Thorpe, J Brazier, M Fitter, M J Campbell, M Roman, S J Walters, and J Nicholl Randomised controlled trial of a short course of traditional acupuncture compared with usual care for persistent non-specific low back pain BMJ 2006; 333: 623 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Alternative must be scientifically brought into mainstream

Rajan TD   (17 September 2006)

Evidence and belief

David Colquhoun   (18 September 2006)

Non-traditional acupuncture is quicker and equally effective

Anthony Campbell   (18 September 2006)

Re: Evidence and belief

Rajan TD   (19 September 2006)

The importance of previous treatment history

Rizaldy Pinzon   (22 September 2006)

Re: Non-traditional acupuncture is quicker and equally effective

Marcus V Ferreira   (22 September 2006)

Re: Re: Evidence and belief in ancient medicines

nicholas D. moore   (23 September 2006)

91% would try it again.

Richard L Davies   (24 September 2006)

Evidence of benefit or bereft of evidence?

Alfred P J Lake   (27 September 2006)

Re: Evidence of benefit or bereft of evidence?

Richard Bartley   (28 September 2006)

Acupuncture or tree-hugging?

Edzard Ernst   (28 September 2006)

Re: Evidence of benefit or bereft of evidence?

Mark C Atkins   (29 September 2006)

Let's look at the facts

David Colquhoun   (29 September 2006)

Re: Acupuncture or tree-hugging?

Richard Bartley   (30 September 2006)

Back Pain: holistic strategy or monotherapy?

Abdul-Ghaaliq Lalkhen, Andrew M. Severn, Consultant in Pain Medicine   (30 September 2006)

Contribution of “traditional acupuncture” may be overstated

Howard H Moffet   (3 October 2006)

Acupuncture: exact treatment effect yet to be determined

Michael Grevitt   (23 October 2006)

Evidence or Desire?

Marieke J. Gieteling, Rogier M. van Rijn, Sten P. Willemsen   (25 October 2006)

Report from BMJ post publication review meeting

Trish Groves   (24 July 2007)

Alternative must be scientifically brought into mainstream 17 September 2006
Rajan TD, Consultant, Skin & Sex Transm Diseases CMPH Medical College, Mumbai, India. 0091-22-66982747 Send response to journal: Re: Alternative must be scientifically brought into mainstream	For decades the lay public have been reading about the beneficial effects of acupuncture for treating painful conditions, for inducing anaesthesia etc. The qualified medical professionals have been hearing 'quacks' claiming success in treating various disorders of the human body. Hordes of people all over the world visit acupuncturists and many of them claim to have got relief of their symptoms from this treatment in disorders where allopathic physicians have thrown their hands up in despair. The current study is a step in the right direction. It is an accepted fact that modern medicine does not have the answer to all diseases. In fact it is absurd to expect mankind to solve ALL health problems using any form of knowledge. At the same time it must be accepted that alternative systems of medicine have answers to some problems that modern medical science is unable to answer. Practitioners of alternative systems of medicine are presumably guilty of not developing a system of documenting the efficacy of their therapies. This is the reason why it could not withstand the onslaught of allopathic medicine which is scientifically analysed, criticised,standardised and simplified for general use. Unfortunately, this process of documentation of allopathy bulldozed the beneficial effects of ayurveda, homoeopathy, siddha, unani, acupuncture etc. Meanwhile unqualified practitioners of alternative systems of medicine took advantage of the confusion which further pushed these ancient systems into a tangle. Therefore, there is an urgent need for all to assimilate the positive features of systems like acupuncture and make use of its beneficial effects. Scientific analysis of these systems need to be undertaken, particularly in chronic disorders, where modern science is yet groping in the dark. A balanced approach of modern medical professionals like us towards systems like acupuncture will help the development of a truly holistic system of healthcare for the benefit of mankind. Competing interests: None declared
Evidence and belief 18 September 2006
David Colquhoun, Professor of Pharmacology University College London, Gower Street, London WC1E 6BT Send response to journal: Re: Evidence and belief	Just one question for Dr Rajan. He admits, quite accurately that advocates of alternative medicine are "guilty of not developing a system of documenting the efficacy of their therapies". That being so, how can he know that "ayurveda, homoeopathy, siddha, unani, acupuncture etc" have "beneficial effects"? You either believe in evidence or you don't. This study gets us little further, for the reasons given in my comment on the cost-effectiveness paper [see below]. Competing interests: None declared
Non-traditional acupuncture is quicker and equally effective 18 September 2006
Anthony Campbell, Former consultant physician, Royal London Homeopathic Hospital Retired Send response to journal: Re: Non-traditional acupuncture is quicker and equally effective	Perhaps the most interesting finding of this trial is the observation that patients' prior belief in acupuncture does not predict the outcome of treatment. From ny own experience I would say that in some cases strong believers actually do worse than those whose attitude is neutral. The acupuncture in the trial was done using the traditional approach, which is time-consuming. Modern Western acupuncture can be performed much more quickly and easily (about 5 minutes' needle insertion in most cases) and the results appear to be quite as good as those of the traditional form. Three out of four recent RCTs (1-4) found no significant difference in effectiveness between traditional acupuncture and "sham" acupuncture (superficial needling at non-acupuncture points). This could indicate either that the beneficial effects of acupuncture are due to the placebo response, or that there is a therapeutic response to needling more or less irrespective of how or where it is done. References: [1] Linde K, Streng A, Jurgens S, Hoppe A, Brinkhouse B, Witt C. Acupuncture for patients with migraine: a randomized controlled trial. JAMA 2005;293:2118-2125. [2] Melchart D, Streng A, Hoppe A, Brinkhaus B, Witt C, Wagenpfeil S. Acupuncture in patients with tension-type headaches: randomised controlled trial. BMJ 2005;331:379-382 [3] Witt C, Brinkhaus B, Jena S, Linde K, Streng A, Wagenpfeil S. Acupuncture in patients with osteoarthritis of the knee: a randomised trial. Lancet 2005;366:136-143 [4] Scharf H-P, Mansmann U, Streitberger K, Witte S, Kramer J, Maier C. Acupuncture and knee osteoarthritis: a three-armed randomised trial. Annals of Internal Medicine 2006;145:12-20 Competing interests: None declared
Re: Evidence and belief 19 September 2006
Rajan TD, Consultant Skin & Sex Transm Diseases CMPH Medical College, Mumbai, India. Tel: 0091-22-66982747 Send response to journal: Re: Re: Evidence and belief	I thank David Colquhoun for his opinion on my response. Traditional systems of medicine are being practised widely in India with good results and they have a large following in various pockets of the country. Likewise, one presumes that other civilisations too have their systems of ancient medicine, many of which have not been given wide coverage. It will be a loss to mankind if such systems are not given a detailed look-in and simply condemned by practitioners of modern medicine Competing interests: None declared
The importance of previous treatment history 22 September 2006
Rizaldy Pinzon, Neurologist Bethesda Hospital Yogyakarta INDONESIA Send response to journal: Re: The importance of previous treatment history	Acupuncture for pain management is very common in my country. The article from Dr. Thomas et al. seems to be very promising for answer the uncertainty of the evidences about acupuncture. I think the previous treatment history is a very important factor. I believe that most of the patients have already taken previous medications. In my experience, patients commonly asked for alternative medicine program when the pain becomes chronic. Drug management is still the mainstay for pain management in my country. The patients who are not satisfied with the pain relief will look for the alternative medicine. The phenomenon is also observed when patients are afraid of the side effects of chronic medications. Competing interests: None declared
Re: Non-traditional acupuncture is quicker and equally effective 22 September 2006
Marcus V Ferreira, MD, private practice Rio de Janeiro, Brasil 22410-002 Send response to journal: Re: Re: Non-traditional acupuncture is quicker and equally effective	Dr. Campbell points to a very polemical issue: which is better? Traditional or non-traditional acupuncture? My opinion is that traditional acupuncture has a huge database comprehending treatments in the past 2.000 years. We should remember that this database was built in a hit and miss procedure, and I think is not wise to waste all that experience. There are some hindrances, of course, among them the translation of a cultural and temporally distant language, but this task is being performed by medical anthropologists like Unschuld, for example. About sham acupuncture, couldn't this be a matter of method? Isn't significant that out of the four RCTs Dr. Campbell cited, three came from the same group (Brinkhaus et al.)? About being quicker, this is not dependent on traditional or non- traditional. Considering electro-acupuncture as a gold standard in pain treatment, there is some evidence that the optimum time of needle stimulation is thirty minutes [1], regardless of which kind of acupuncture is being practiced, traditional or not. References: [1] Hamza MA et al., Effect of the duration of electrical stimulation on the analgesic response in patients with low back pain, Anesthesiology, 91:1622-7, 1999 Competing interests: None declared
Re: Re: Evidence and belief in ancient medicines 23 September 2006
nicholas D. moore, Professor of clinical pharmacology Université Victor Segalen, 33076 Bordeaux Send response to journal: Re: Re: Re: Evidence and belief in ancient medicines	Indeed one finds ancient medicines everywhere, that evolve over time, and like every evolutionary creature, some branches thrive, other wither and died, and some survive in ecological niches. Ancient european medicine has indeed evolved in what is commonly called modern medicine. Some fossil remnants of older species that have not really evolved can be found in homoeopathy, or in central asian traditional medicines (an evolutionary pocket isolated by the closing of the silk road when european medicine started changing), and some can have quite a large following. There are still folk medicines in various parts of europe, though I am not sure they qualify as ancient medicines... All these work quite well until they come under scientific scrutiny, which shows that they mainly work through the practitioner and patient's belief in their efficacy (belief and faith are powerful medicines indeed). 18th century medicine in europe certainly had good results even though it has been generally shown, I understand, that the body is not governed by humours secreted by the liver. But then maybe that traditional approach should be revived and reinstated. After all, like evolution, physiology is but a theory, and no one has actually seen a germ with their own eyes, have they? Good luck Nicholas Moore Competing interests: I tend to be a pharmacologist and believe in scientific evidence, and therefore in mainstream physiology and in evolution.
91% would try it again. 24 September 2006
Richard L Davies, General Practitioner 703 Leeds and Bradford Road, Stanningley, Pudsey, Leeds LS28 6PE Send response to journal: Re: 91% would try it again.	In their randomised controlled trial of acupunture for back pain K J Thomas et al state 91% of the acupuncture group would try it again (1). Do the authors know if any of the acupuncture group went on to fund extra treatment in the period between the inital short course and the final assessment? In my experience when patients with chronic medical problems find a treatment that works, they are often keen to continue it. (1) Thomas K J, Macpherson H, Thorpe J, Brazier J, Fitter M, Campbell M J, Roman M, Walters S J, Nicholl J. Randomised controlled trial of a short course of traditional acupunture compared with usual care for persistent non-specific low back pain. BMJ 2006;333:623-6(23 September) Competing interests: None declared
Evidence of benefit or bereft of evidence? 27 September 2006
Alfred P J Lake, Consultant in Anaesthesia and Pain Management Glan Clwyd Hospital, Rhyl LL18 5UJ Send response to journal: Re: Evidence of benefit or bereft of evidence?	Acupuncture may, or may not, be of benefit in persistent non-specific low back pain but this study, unfortunately, goes no way to providing an answer, contributes little, and it is somewhat surprising to find it in a prestigious journal such as the BMJ. We must consider it in conjunction with the editorial (1) and accompanying research paper (2) in the same issue. In this day and age we must, surely, move beyond considering important ‘relatively cost-effective’ interventions and rate a study as valuable just because it is ‘well conducted’ and ‘statistically significant’. How does a ‘modest’ reduction warrant an editorial? It is also particularly important that we do not ignore the opportunity cost of spending even small amounts on questionable interventions; it all adds up, and we know that the placebo effect is strong, particularly with respect to pain management. The design of the study includes a huge number of potentially confounding variables which might explain why it took so long to get it together for publication as the trial ran between 1999 and 2001. Just 241 very different patients (age 18-65 with pain of 4-52 weeks’ duration) very thinly spread (16 {or 18} practices and 39 {or 43} GPs) and no mention of Yellow Flag evaluation which we know to be of major importance for this group of patients (3). The 6 acupuncturists may have been registered with an appropriate body but they were delivering individual (and individualised) care nonetheless. The components of acupuncture treatment identified (individualised treatment, interaction effect, good therapeutic relationship with protected time and attention, relaxing experience) would all serve to boost the benefits for patients experiencing chronic pain. Also, further NHS acupuncture was received by patients in both groups and, in addition, 19 patients in the acupuncture arm of the study purchased more privately; they were obviously positive about its benefits. Patients not randomised to the acupuncture arm received NHS treatment according to their general practitioner’s assessment of need i.e. 39 (or 43) different assessments and the interventions themselves were not standardised and comprised a mix. Weak evidence of an effect, it seems, was found; in reality differences failed to reach statistical significance in most analyses with just a small (perhaps clinically insignificant) difference in SF36 at 24 months and no evidence of functional improvement. Evidence is the data on which a judgement or conclusion may be based or by which proof or probability may be established; if this study is ‘best evidence’ then it just confirms that we can now make statistics work to produce a ‘positive’ result from almost any data. Acupuncture may well be of worthwhile benefit for patients with persistent non-specific low back pain but this study does not support such a contention. Best evidence? Not this, do it again and do it better! 1. Wonderling D. Acupuncture in mainstream health care. BMJ 2006;333:611-2 2. Ratcliffe J, Thomas KJ, MacPherson H, Brazier J. A randomised controlled trial of acupuncture care for persistent low back pain: cost effectiveness analysis. BMJ 2006;333:626-8 3. New Zealand Acute Back Pain Guide. October 2004 edition. http://www.nzgg.org.nz/guidelines/0072/acc1038_col.pdf (accessed 26.09.06) Competing interests: None declared
Re: Evidence of benefit or bereft of evidence? 28 September 2006
Richard Bartley, Physiotherapist Denbigh Infirmary Send response to journal: Re: Re: Evidence of benefit or bereft of evidence?	I agree. Time and time again I read studies on back pain intervention where the outcome tools used have little to do with the price of fish. Acupuncture delivers a sensory stimulation with a frequency around 2Hz. Stimulating needles placed in the right dermatone may release opiod compounds in the CNS, resulting in temporary pain relief. Add a 90 - 130 Hz electric current and you might also be able to fiddle around with the pain-gate mechanism. And that's fine. What does it matter what the patient's ODI or SF36 score is in a year's time? The patient comes in, gets some temporary pain relief, then goes away. The patient gets a few weeks pain relief, which from his or her perspective is good value for money. Of course from a physiotherapists' point of view, I would be keen to recommend that the patient gets to grip with this self-limiting but rarely dangerous condition, by making a few simple life style changes and getting a bit fitter. Competing interests: None declared
Acupuncture or tree-hugging? 28 September 2006
Edzard Ernst, Director Complementary Medicine, 25 Victoria Park Road, Exeter, EX2 4NT Send response to journal: Re: Acupuncture or tree-hugging?	The papers by Thomas et al. [1] and Ratcliffe et al. [2] suggesting that acupuncture is an effective and cost-effective therapy for low back pain raise several important issues. The authors fail to mention that these data have already been published 2 years ago[3], and in 1999 the authors published their protocol [4]. Comparing this protocol [4] and the published account [1] of the trial, one stumbles over several contradictions. For instance, the BMJ article states that the aim of the study was “to determine whether acupuncture improves longer term outcomes” [1], however, the protocol tells us that the aim was to find out whether the offer of acupuncture improves pain (regardless of uptake of this offer) – perhaps only a small discrepancy but, I think, an important one. The protocol also does not mention one primary outcome measure but three, two of which turned out to show no significant effects. Crucially, the protocol mentions only 12 months of follow-up. The 24 months follow-up is thus a post hoc addition. Interestingly, the results were significant only at that time point. Thus, according to the published protocol [4], this study failed to generate a positive finding – a fact that is certainly not expressed in the BMJ articles [1], [2]. The trial did not control for placebo-effects and the authors avoid discussing in detail whether the benefits they report are due to a placebo response. They fail to mention, for instance, two large German trials, which did attempt to control for placebo effects, found no difference between real and minimal acupuncture[5]. Non-penetrating placebo needles have recently been developed and validated which can answer the placebo question. An overview of the 13 clinical trials using such placebo controls suggests that acupuncture may well be little more than a placebo[6]. In this situation, pragmatic trials such as this of Thomas et al[1] may seriously mislead healthcare policy, and even the most rigorous cost-analysis may only demonstrate the cost-effectiveness of placebo for a self-limiting condition. To put it bluntly, hugging a tree may even be more cost-effective (and safer) than acupuncture. Reference List 1. Thomas KJ, MacPherson H, Thorpe L, Brazier J, Fitter M, Campbell MJ et al. Randomised controlled trial of a short course of traditional acupuncture compared with usual care for persistent non-specific low back pain. BMJ 2006;333:623. 2. Ratcliffe J, Thomas KJ, MacPherson H, Brazier J. A randomised controlled trial of acupuncture care for persistent low back pain: cost effectiveness analysis. BMJ 2006;333:626. 3. Thomas KJ, MacPherson H, Ratcliffe J, Thorpe L, Brazier J, Campbell M et al. Longer term clinical and economic benefits with chronic low back pain. Health Technology Assessment 2005;9. 4. Thomas KJ, Fitter M, Brazier J, MacPherson H, Campbell M, Nicholl JP et al. Longer term clinical and economic benefits of offering acupuncturre to patients with chronic low back pain assessed as suitable for primary care. Comp Ther Med 1999;7:91-100. 5. Bäcker M, Tao I, Dobos GJ. Akupunktur - quo vadis? Dtsch Med Wochenschr 2006;131:506-11. 6. Ernst E. Acupuncture - a critical analysis. J Intern Med 2006;259:125-37. Competing interests: None declared
Re: Evidence of benefit or bereft of evidence? 29 September 2006
Mark C Atkins, Consultant Virologist Hammersmith Hospital W12 0HS Send response to journal: Re: Re: Evidence of benefit or bereft of evidence?	Dr Lake makes some valid points. This study failed to show a statistically significant benefit if favour of acupuncture with respect to the primary end-point. If this was a pharmaceutical product I doubt anyone would be rushing to refer patients to be treated with it. It certainly wouldn't get a license. As for the 24 month results, there is a lot of missing data and any analysis has to be interpreted with great caution because of this. A more rigorous intent-to treat analysis would have included the missing data points as failures. Competing interests: None declared
Let's look at the facts 29 September 2006
David Colquhoun, Professor of Pharmacology University College London, Gower Street, London WC1E 6BT Send response to journal: Re: Let's look at the facts	Following your publication of the paper by Thomas et al. the Daily Express trumpeted "Acupuncture is best remedy for backpain". Let's just look at the facts. (1) The effects seen in this trial were tiny. (a) Functional disability was not improved (b) Three different pain scales were used. On two of these (Oswestry pain disability index and the McGill present pain intensity measure) there was no significant difference between acupuncture and control at any stage. On the third (the SF-36 scale) there was no significant effect at 12 months either (95% confidence interval -0.2 to 11.4) At 24 months there was a significant effect (just), limits (2.8 to 13.2).but this means there is a chance not much smaller than 5% that there are no effects at all, even on the SF-36 scale. This is hardly a very secure result. (c) Even if the effect is quite genuine on the one scale where an effect was found (SF-36), it is a very small effect (8 points), below the authors' own modest aims. "our study was designed to detect a larger difference of 10 points, which was not achieved at either 12 or 24 months." (2) Even if the effect is real, and even if such a small effect is considered worthwhile, there isn't the slightest reason to attribute the effects to acupuncture per se. Because it was a "pragmatic trial", i.e. had no appropriate controls, it is entirely plausible that all it showed that somebody armed with needles and talking gobbledygook about meridians, in the setting of a private clinic, can elicit a slightly bigger placebo effect than that elicited by the interview in the NHS clinic. The paper says this (just) "An open pragmatic trial avoids the potential problems associated with using sham acupuncture as the control.26 Such a design may, however, be vulnerable to confounding or bias owing to prior patient beliefs about how acupuncture might help, especially when using subjectively assessed outcome measures, such as perceived pain." What are the problems associated with the use of sham acupuncture? The only problem that I can see is that it might show that acupuncture is nothing but placebo! The rest of the sentence is just a convoluted and obscure way of saying this. (3) The points I make above seem to me to mean that this is just yet another inconclusive trial, even at the "pragmatic" level. Because, within its own (inadequate) rules it appears to have been carried out quite well, I'm not complaining about your decision to publish it. But I am complaining about the press release that accompanied it. It cast an indecisive result in a glowing light, and accordingly got publicity for the BMJ in the newspapers, who treated it as though it was a strong endorsement that "acupuncture works". Of course you could blame this misinterpretation on the journalists, but it is not helped by misleading press releases. In particular you failed to point out that a pragmatic trial cannot, by its nature, give any evidence at all about whether acupuncture works better than placebo. Since that was specifies by the House of Lords report (2000) as the first priority to be established before CAM could be taken seriously, it is arguable that you should not be publishing pragmatic trials at all until such time that that hurdle is passed. If you must publish them, the surely you should point out the limitations of such trials to journalists. In this case, I should have thought it very dubious that such a small and indecisive result deserved a press release at all. Are you sure that the purpose of the release was to inform the public about medicine, as opposed to being a chance to get the BMJ into the papers (knowing that they will lap up anything to do with CAM)? Competing interests: None declared
Re: Acupuncture or tree-hugging? 30 September 2006
Richard Bartley, Physiotherapist Denbigh Infirmary LL16 3ES Send response to journal: Re: Re: Acupuncture or tree-hugging?	Tonight I shall conduct a clinical trial (n=2) on the effects of tree hugging on back pain. My wife and I, both sufferers, will complete the SF- 36, the Oswestry Disability Index, McGill Pain Questionnaire and a patient satisfaction survey. Once those are out of the way I will go down the woods for half an hour of intensive tree hugging. My wife will stay at home and take an aspirin. I shall repeat this therapy ten times over the next three weeks. I am confident that I will find the therapy deeply satisfying and far more beneficial to me than popping pills. Over the next two years, I’ll probably sneak off to an osteopath from time to time without letting anybody know, take up jogging and badminton, lose some weight, climb a mountain, get a new job so I don’t have to lift anything heavy anymore, go on holiday to Jamaica and get mugged, develop a new illness that is more terrifying than back pain and to top it all off, get divorced. Then when the two years is up, I shall complete my SF-36, Oswestry Disability Score, McGill Pain Questionnaire and the patient satisfaction survey again. My wife won't be around to complete hers as by then she will have run off with the new gym instructor, but I'll record her in the study as natural attrition. To heck with insignificant McGill and ODI scores, if I get an intervention effect of 5.6 points on the primary outcome SF-36 pain dimension, I shall be cock-a-hoop. Competing interests: None declared
Back Pain: holistic strategy or monotherapy? 30 September 2006
Abdul-Ghaaliq Lalkhen, Pain Fellow North West School of Anaesthesia, Andrew M. Severn, Consultant in Pain Medicine Send response to journal: Re: Back Pain: holistic strategy or monotherapy?	Sir, Your articles and editorial on the cost effectiveness of acupuncture – a ‘traditional therapy’- for the treatment of back pain - an affliction of industrial societies - raises important practical, philosophical and ethical points The standardization of treatment is poor. Therapists are allowed to choose the ‘content and number of treatments’ . Some 18% of patients chose acupuncture over and above that deemed to be appropriate by the therapist and 6% recovered before treatment was started. There is no standardization of ‘usual treatment’ in the control group. There are many guidelines on the conservative management of back pain which could have been followed to ensure a uniform cohort of controls. The fact that the groups actually differed before the trial in an important outcome variable of back pain management - the ability to work - is a serious lapse. The use of additional therapies in the experimental group makes the specific contribution of acupuncture even more difficult to identify It is encouraging that the authors recognise principles of good practice of managing patients with distressing and difficult pain. These include: ‘the practitioner’s skills at developing good therapeutic benefit…..protected time and attention from the practitioner’ as part of the recipe for success. However, it is unclear whether the ‘usual care’ offered to patients in the control group was dispensed with the same degree of attention. Whether any attempt in ‘usual care’ was made to address the ‘concerns about back pain’ that were mentioned in the acupuncture group is likewise unclear, but such discussion should be a feature of the ‘usual’ modern management of back pain by whomever it is practised. This paper shows that ‘trained acupuncturists’ are able, like many other professionals, to deliver a package of physical and psychological interventions for back pain sufferers. However, the precise contribution of acupuncture to outcome remains as elusive as ever. The estimated cost of £17 per single treatment does not seem excessive, but it can also provide approximately 25 minutes of a consultant’s salary in an outpatient clinic or just over an hour for a senior physiotherapist : time that could be spent on to the comprehensive assessment of one more new patient or the review of two others. There is a conflict of interest here that has been not disclosed. Two researchers belong to an organization whose stated purpose is ‘.. to bring the traditional Chinese system of acupuncture more centrally into the national health system’ . The failure to justify the exclusion of any patient over the age of 65 makes it difficult to apply the conclusions to NHS practice generally (1). These failures are sufficiently serious to make us question whether this work should have been published. Reference 1) Unjustified exclusion of elderly people from studies submitted to research ethics committee to approval: descriptive study Bayer A, Tad W British Medical Journal 2000, 321,992-993 Competing interests: None declared
Contribution of “traditional acupuncture” may be overstated 3 October 2006
Howard H Moffet, Senior Research Manager Kaiser Permanente - Division of Research, Oakland, CA USA 94612 Send response to journal: Re: Contribution of “traditional acupuncture” may be overstated	While it is gratifying to read of a potentially beneficial treatment for patients with low back pain, this paper did not describe the study intervention in any useful way, failing even to describe which “tradition” of acupuncture practice was used or whether the “content” of the intervention was limited to acupuncture needling. In this poorly controlled trial, the authors “evaluated a package of care and cannot isolate the components of acupuncture treatment that are associated with the outcomes observed in the acupuncture group.” The authors suggested five alternative explanations for the results, including “individualizing of treatment,” by which they may mean that there was no fixed treatment protocol. Under the rubric of “traditional acupuncture,” patients may have each received needles at unique combinations of points, stimulated at various depths and with various intensities, perhaps varying at each visit. While the art of traditional acupuncture may allow a thousand subtle variations in practice, there is little evidence that the variations make any important difference. The latter is supported by their alternative explanation of a “widely reported relaxing experience of the treatment itself,” which suggests that any needling may induce a beneficial state of relaxation. The “traditional acupuncture” approach to point selection and needling technique may contribute more to obfuscation than insight. This poorly-controlled study provides little evidence of acupuncture’s efficacy for low-back pain and the contribution of “traditional acupuncture” methodology remains dubious. The lack of information about the intervention makes the study impossible to reproduce, experimentally or in practice, and is particularly disappointing in a scientific journal. Competing interests: None declared
Acupuncture: exact treatment effect yet to be determined 23 October 2006
Michael Grevitt, Consultant Spinal Surgeon Queen's Medical Centre, Nottingham Send response to journal: Re: Acupuncture: exact treatment effect yet to be determined	This study claims a small benefit for acupuncture in persistent non- specific low back pain after 24 months. This conclusion however must be limited with several caveats that are not discussed in the paper. The patient population is skewed towards the acute spectrum of low- back pain (as evidenced by the modest disability compared with a chronic back pain population) [1] There is evidence that the outcomes of non-surgical interventions in back pain are dose-dependent i.e. the more intense and frequent the exposure the better the outcome. The disparity in treatment regimens of the two groups was likely to favour the acupuncture patients. Despite this the only significant difference between the groups was the SF36 bodily pain score at 24 months (and this difference only just exceeds published minimally important clinical difference) [3]. This small benefit is not mirrored in the more specific McGill pain questionnaire scores at any time point in the follow-up period. Similarly, there was no difference in the Oswestry disability index scores nor was there any functional improvement in either group. The epidemiology of acute back pain suggests that spontaneous resolution occurs in the majority although this may take longer than was hitherto believed [4]. This study confirms the phenomena of regression to the mean in non-specific low back pain. The methodology does not allow exact characterisation of the effect of acupuncture. Any claimed benefit (which is open to statistical challenge), may be the result of the complex interaction between patient and acupuncturist rather than the needle therapy. In summary, this study does nothing to refute the conclusion of van Tulder et al [5]; there is no evidence that acupuncture is better than sham or placebo. 1. Fairbank J, Frost H, Wilson-MacDonald J, et al.Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial. BMJ. 2005 May 28;330(7502):1233. Epub 2005 May 23. 2. Guzmán J, Esmail R, Karjalainen K, Malmivaara A, Irvin E, Bombardier C. Multidisciplinary Bio-Psycho-Social Rehabilitation for Chronic Low Back Pain (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Oxford 3. Hagg O, Fritzell P, Nordwall A; Swedish Lumbar Spine Study Group. The clinical importance of changes in outcome scores after treatment for chronic low back pain. Eur Spine J. 2003 Feb;12(1):12-20. Epub 2002 Oct 24. 4. Croft PR, Macfarlane GJ, Papageorgiou AC, Thomas E, Silman AJOutcome of low back pain in general practice: a prospective study. BMJ. 1998 May 2;316(7141):1356-9. 5. van Tulder MW, Cherkin DC, Berman B, Lao L, Koes BW. Acupuncture for low back pain (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Oxford Competing interests: None declared
Evidence or Desire? 25 October 2006
Marieke J. Gieteling, GP trainee, PhD student Department of General Practice, Erasmus Medical Centre, 3000 CA Rotterdam, The Netherlands, Rogier M. van Rijn, Sten P. Willemsen Send response to journal: Re: Evidence or Desire?	Dear editor, With interest we read the paper by Thomas et al. (2006) reporting on their pragmatic trial of acupuncture in low back pain patients, recruited in general practice. ¹ In general the trial is nicely designed and conducted. Still, we would like to raise several issues, which may affect the interpretation of the findings. First, we wonder why the SF-36 pain score was chosen as primary outcome measure. This is a rather unusual and not a recommended instrument for low back pain.² A more adequate outcome measure might have led to a different outcome. Secondly, with a loss to follow-up of 24%, we feel that important information about the reasons for dropping out is missing. The authors themselves write: “the effect of the missing data is unknown”, but the dropout was reported to be selective. This high and selective dropout rate affects the internal validity and the generalizability of the study. Third, the authors adjusted for the clustering effect of the acupuncturists, but not for a possible similar effect of the GPs. As a result, confidence intervals may be spuriously narrow. Finally, the authors’ conclusion that there was “weak evidence of an effect of acupuncture at 12 months” is not justified since at that time point no statistically significant result was obtained. We feel that the proper conclusion should be that there was no evidence of an effect at 12 months. References 1. Thomas K, MacPherson H, Thorpe L, Brazier J, Fitter M, Campbell M, et al. Randomised controlled trial of a short course of traditional acupuncture compared with usual care for persistent non-specific low back pain. BMJ 2006;333:623-8. (15 September.) 2. Deyo R, Battie M, Beurskens A, Bombardier C, Croft P, Koes B, et al. Outcome measures for low back pain research. A proposal for standardized use. Spine 1998;23:2003-13. Competing interests: None declared
Report from BMJ post publication review meeting 24 July 2007
Trish Groves, Deputy editor, BMJ BMJ, BMA House, Tavistock Square, London WC1H 9JR Send response to journal: Re: Report from BMJ post publication review meeting	At the BMJ we have two forms of continuing appraisal of articles we have published: Rapid Responses and the annual Post Publication Review Meeting. At the review meeting statistical advisers and BMJ research editors discuss critically how we handled articles which have proved particularly controversial. Here are the conclusions of our review of our handling of this paper: K J Thomas, H MacPherson, L Thorpe, J Brazier, M Fitter, M J Campbell, M Roman, S J Walters, and J Nicholl Randomised controlled trial of a short course of traditional acupuncture compared with usual care for persistent non-specific low back pain BMJ 2006;333:623 The study was sound, but the effect size and its clinical meaning should have been reported and discussed more clearly. The published study differed from the protocol, which had been published previously on the Health Technology Assessment website. This protocol deviation was valid but not adequately described. Lessons learned and actions we're taking: The BMJ will clarify advice to authors on how to report effect sizes and related uncertainty. When appropriate, we will ask authors to report that they "found no convincing evidence of an effect". We will ask authors to discuss the clinical importance of main outcomes in such studies as well as effect size. We will ask authors to describe and explain fully any deviations from protocol. Competing interests: I am senior research editor at the BMJ