In praise of chiropractic?
Posted by apgaylard on July 3, 2009
The Journal of Health Services Research and Policy has just published a ‘perspective’ piece by Professor Alan Breen of the Anglo-European College of Chiropractic in which he, unsurprisingly, praises Chiropractic (Breen, 2009).
The concluding paragraph is quite amazing:
“To portray only part of the relevant information in a critique is itself pseudoscience, yet strong ontological commitment to only part of the knowledge base seems often to be the stance taken to contest the scientific basis of Chiropractic. Rather like psychiatry, debates about musculoskeletal practice need to go beyond positivist science and recognize that we have entered a more democratic and inclusive, post-normal age.” (Breen, 2009)
In this he seems to be committing himself to evaluate Chiropractic in the light of all the relevant information. Let’s see if he does.
First, a bit of philosophy. His appeal to go beyond “positivist science” is puzzling. It is unclear whether he is assuming that contemporary science as a whole is “positivist”; or that it’s just the science that criticises Chiropractic. Equally, it could be that Breen is using “positivist” as code for “science I don’t like”.
In any event, there seems to be an assumption here that scientists with a different philosophical perspective would come to more positive conclusions about Chiropractic.
None of these is a particularly convincing position. At its heart, positivism asserts that testability and meaning are essentially the same thing. This philosophical concept was first defined by Auguste Comte (1798 – 1857); in the early years of the 20th century it was developed into a stricter and more formal version – logical positivism – by the Vienna Circle, as an attempt to define a philosophical structure for science. In its original form logical positivism asserted meaning through verification – the proving or confirming of the truth of a statement. Although this strong verificationalism didn’t last long – Karl Popper is perhaps the best known critic of applying this idea to science – logical positivism was perhaps the dominant philosophical explanation of science for the first half of the twentieth century.
Although heavily modified over the years, it turns out that science is a much too complex and messy human endeavour to be defined by such a philosophical straight jacket like that proposed by the positivists.
Contrary to what Breen appears to be saying, science has actually gone beyond “positivist science”. Another problem with his assertion of positivism being the problem is that it is a philosophy that has nothing to say about statistical reasoning. In its article on positivism, the Encyclopædia Britannica (2009) reminds us that:
“[…] The significant role of statistical (or probabilistic) explanations in most modern sciences, for example, is receiving increasingly sophisticated analyses.”
And, of course, the kind of science that usually upsets CAM advocates is that based on statistical inference. Medical trails of efficacy of therapies usually boil down to an assessment of whether the differences between a ‘control’ and ‘treatment’ group is statistically significant, that is having a pre-defined probability that the result isn’t just a fluke. Arguments against Chiropractic made from statistical reasoning have nothing to do with positivism.
This is made even clearer by thinking about the statistical practise of seeking evidence against the null hypothesis (that the outcomes for ‘treatment’ and ‘control’ groups are not different). In the words of the father of statistical tests of significance:
“Every experiment may be said to exist only to give the facts the chance of disproving the null hypothesis.” (Fisher, 1935).
This kind of testing shares a philosophical heritage with the falsificationism of Popper, not the Positivists.
This also kills off the idea that a positivistic approach is responsible for critiques of Chiropractic: one of the main strands of evidence against its efficacy comes from Randomised Control Trails (RCTs). These are philosophically compatible with the many different philosophical positions within science. All they require is commitment to the fundamental shared scientific values of fair testing and the avoidance of self-deception.
An appeal to politics?*
The closing appeal of the article to, “recognize that we have entered a more democratic and inclusive, post-normal age” when assessing Chiropractic practise for musculoskeletal disorders is a red herring. Ideas around democracy and inclusivity have no bearing on whether a medical treatment works or not. Otherwise NGOs would be wasting their time trying to get medical aid into non-democratic societies during times of trouble. Neither does a lack of democracy in China mean that acupuncture works or that scientific medicine does not: it’s not a relevant factor.
The bottom line is that human social and political constructs may have the power to influence the outcome of some scientific studies; but they cannot alter whether subluxations exist (they do not) or whether Chiropractic treatment works (as opposed to the expectations and beliefs of chiropractors and their customers making them think it does). In fact, scientific values and methods have been developed with the aim of minimising these influences.
Where’s the beef?
So, empty post-modern posturing aside, is there anything substantial in Breen’s attempt to praise Chiropractic?
Well, his apologia focuses on back pain specifically. So this is not a defence of Chiropractic practise in general, but a more limited attempt to praise Chiropractic for its approach to the management of back pain.
On this topic he notes that, “Manual therapies have been consistently recommended in evidence-based guidelines in the UK since 1994.” This is true**, but whilst Chiropractic is (by definition) a manual therapy, not all manual therapies are Chiropractic.
Breen does not offer any trial data to support the notion that Chiropractic is effective for the treatment of back pain. In fact, he does not attempt an even-handed review of the evidence. After accusing critics of Chiropractic of pseudoscience by virtue of an alleged, “strong ontological commitment to only part of the knowledge base” he demonstrates the same tendency.
This can be seen in his assertion that:
“[…] the consistent evidence is that manual therapies, (unlike most interventions that have been scientifically evaluated) are helpful and generally produce moderate but significant and sustained improvement for back pain in populations. In addition, and contrary to much critical commentary on Chiropractic, they are not by any means all that chiropractors do.” (Breen, 2009)
A quick review of the evidence
No citations are provided by the author to back up these claims. To try and demonstrate a strong ontological commitment to the complete knowledge base I searched the Cochrane Library***and assumed that as the article was trying to praise Chiropractic, then Chiropractic was the issue – rather than other manual therapies.
However, it is worth mentioning in passening that a review of spinal manipulative therapies for low-back pain (Assendelft et al., 2004) concluded that:
“[…] For patients with acute low-back pain, spinal manipulative therapy was superior only to sham therapy […] or therapies judged to be ineffective or even harmful. Spinal manipulative therapy had no statistically or clinically significant advantage over general practitioner care, analgesics, physical therapy, exercises, or back school. Results for patients with chronic low-back pain were similar. Radiation of pain, study quality, profession of manipulator, and use of manipulation alone or in combination with other therapies did not affect these results.”
And that:
“There is no evidence that spinal manipulative therapy is superior to other standard treatments for patients with acute or chronic low-back pain.”
This indicates that spinal manipulation generally offers no advantage over, say, general practitioner care. This raises the question: if a patient is already under the care of their general practitioner – why try spinal manipulation?
Anyway, that’s spinal manipulation in general; what about Chiropractic in particular?
There are currently no relevant Cochrane Reviews, though French et al (2005) have a review protocol listed. My search found eight other reviews. Starting with those examining efficacy for back-pain, Ernst and Canter (2006) concluded that:
“Collectively these data do not demonstrate that spinal manipulation is an effective intervention for any condition. Given the possibility of adverse effects, this review does not suggest that spinal manipulation is a recommendable treatment.”
Ferreira et al (2003) reviewed the evidence for the “Efficacy of spinal manipulative therapy for low back pain of less than three months’ duration”. Here is part of their abstract:
“[…] Thirty-four papers (27 trials) met the inclusion criteria. Three small studies showed spinal manipulative therapy produces better outcomes than placebo therapy or no treatment for nonspecific low back pain of less than 3 months duration. The effects are, however, small. The findings of individual studies suggest that spinal manipulative therapy also seems to be more effective than massage and short wave therapy. It is not clear if spinal manipulative therapy is more effective than exercise, usual physiotherapy, or medical care in the first 4 weeks of treatment. CONCLUSIONS: Spinal manipulative therapy produces slightly better outcomes than placebo therapy, no treatment, massage, and short wave therapy for nonspecific low back pain of less than 3 months duration. Spinal manipulative therapy, exercise, usual physiotherapy, and medical care appear to produce similar outcomes in the first 4 weeks of treatment.”
Pengel et al (2002) looked at the evidence base of a variety of interventions for sub-acute low back pain. They found few studies and that these could provide little evidence due to methodological flaws. They also commented on, “[…] the lack of a standard definition for sub-acute low back pain.” The effect of spinal manipulative therapy did not reach statistical significance in the trials analysed.
These reviews are consistent with the work of Assendelft et al (1996) who noted:
“Eight RCTs were identified. All RCTs had serious flaws in their design, execution and reporting. Because of the great variety of outcome measures and follow-up timing, there was insufficient data to enable statistical pooling of the RCTs. A narrative review, however, did not provide convincing evidence for the effectiveness of Chiropractic for acute or chronic low back pain.”
Evens and Richards (1996) concluded that, “Research to date has been insufficiently rigorous to give clear indications of the value of treatment for non-specific [low back-pain] patients. No treatment has been shown beyond doubt to be effective.”
So, it would appear that there is little evidence that Chiropractic can really help with back pain. If it does, then the effect is most likely “small” and similar to a number of other interventions. Neither does it appear that the evidence has strengthened over time.
One of the other reviews, Gemmell and Miller (2006) attempted to, “critically appraise the literature that directly compared manipulation, mobilisation and the Activator instrument for non-specific neck pain.” Of the five studies they found, “The methodological quality was mostly poor. Findings from the studies were mixed and no one therapy was shown to be more effective than the others.” And finally, “Further high quality research has to be done before a recommendation can be made as to the most effective manual method for non-specific neck pain.” This appears to indicate that even if Chiropractic was of some use for this condition, there is no evidence to suggest what sort of Chiropractic treatment should be recommended.
Finally, a review by Oliphant (2004) concluded that Chiropractic is a safe treatment for lumbar disk herniations.
Looking at the evidence from reviews, many of them published in the Chiropractic literature, shows that the supporting evidence for the use of Chiropractic in treating back pain is, to be polite, marginal. This evidence is not compatible with Breen’s positive summary.
What are chiropractors good for?
The argument for giving chiropractors credit for their use of non-Chiropractic interventions seems odd to me. First, the trials published in the Chiropractic literature focus almost exclusively on spinal manipulation. So, it would seem to me that if chiropractors wanted the focus to be on other approaches then they should undertake and publish research on these other approaches. Also, survey data from UK chiropractors (GCC, 2004) also shows that spinal manipulation is overwhelmingly what chiropractors do in their clinics.
Finally, if non-Chiropractic interventions are so important, why go to a chiropractor in the first place? This question is also relevant to assessing Breen’s main positive argument: that chiropractors have a long history of doing the sorts of things now recommended for the treatment of back pain.
It may well be true that, “chiropractors and osteopaths were using today’s evidence-based approach in the 1970s” (Breen, 2009). However, the evidence appears to suggest that the Chiropractic component – their one distinctive contribution – is superfluous.
It is also interesting to see Breen advocating, “a biopsychosocial approach”. This seems very sensible and is exactly what good conventional medical care seeks to do****. Again, the role of spinal manipulation would seem to be superfluous within such a framework.
As for a UK parliamentary committee report concluding that, “the practice of osteopathy and Chiropractic were not in conflict with conventional medical practice and that there was evidence of effectiveness of their interventions” I would say that politicians are not a relevant authority in a debate about the safety and efficacy of Chiropractic.
They might have fairly reviewed the data, but there again the reviews listed in the Cochrane Library would suggest that they did not. Neither have I seen anything that would make me inclined to agree with Breen’s unsupported claim that the evidence of efficacy, “has since strengthened”: this also seems at variance with the reviews that I’ve found.
Given recent events, the claim that, “The General Chiropractic and Osteopathic Councils have since proved their mettle in the exacting world of health care regulation” provides (at least on the Chiropractic side of the debate) a moment of pure comedy.
Much of Breen’s argument is about the perceived unfairness of critics who argue that Chiropractic is unsafe.
“[…] members of both professions [Chiropractic and Osteopathy] have looked on in bewilderment at regular, frequent commentaries suggesting that Chiropractic (but not osteopathy) does more harm than good.
The suggestion that Chiropractic does harm is specious.” (Breen, 2009)
Breen seems to miss the point that for Chiropractic to do more harm than good does not require it to do very much harm at all, as the evidence is that it does very little (if any) good!
It is true that estimates for the rate of serious damage done by chiropractors to their patients are low. However, in a risk-to-benefit ratio the Chiropractic denominator is very small. Therefore it is right to be concerned about low rates of incidence of serious complications: particularly when some of the alternatives have even lower risks for the same benefit (return to work or exercise, for example).
The assertion that the suggestion that Chiropractic does harm is “specious” is a bit intemperate. Yes, much of the evidence shows correlation – which does not necessarily imply causation. However, in the presence of a plausible mechanism (high-velocity thrusts, for example) it would be rash to ignore such evidence.
Actually, Breen (2002) published an article entitled, “Manipulation of the neck and stroke: time for more rigorous evidence.” This would seem to be a more reasonable position to take. Still, even in that piece he seems to be making the argument that as correlation is not causation, then chiropractors should keep on treating patients.
This is attempting to argue about risks, without balancing this against the “small” potential benefit or the relative risks of other interventions.
It is really peculiar how often people accuse others of failings that they go on to demonstrate. Breen provides an excellent example of this human foible; concluding that, “strong ontological commitment to only part of the knowledge base seems often to be the stance taken to contest the scientific basis of Chiropractic” at the end of an argument where he neglects to mention any trials that test the efficacy of Chiropractic for any condition. Moreover, the article does not cite a single negative study or review. Instead, politicians and committees are the preferred source of authority.
Again, a preference for one side of the knowledge base is seen in the author’s focus on risks that are not balanced against benefits. Having had a look for the evidence, it appears to me that in an evidence-based biopsychosocial model for the treatment of musculoskeletal complaints: Chiropractic is, at best, unnecessary.
Disclaimer
Please remember that I’m not offering you any medical advice! If you need that consult a proper doctor.
Notes
*The excellent Evidence Matters blog has a much more astute analysis of this strand of Breen’s argument: Alan Breen in praise of chiropractic and democracy.
**Partly, are they really evidence based if they include Chiropractic?
*** I searched for Chiropractic AND “back pain” on 2nd July 2009.
****See here for Ben Goldacre getting it trouble for mentioning “psychosocial risk factors” when discussing back-pain!
Consulting the Profession: A Survey of UK chiropractors, 2004. General Chiropractic Council. 44 Wicklow Street London WC1X 9HL. page 39. Available from: http://www.gcc-uk.org/files/link_file/ConsultTheProfession.pdf.
Positivism. (2009). In Encyclopædia Britannica. Retrieved July 01, 2009, from Encyclopædia Britannica Online: http://www.britannica.com/EBchecked/topic/471865/Positivism
Assendelft WJJ, Morton SC, Yu EI, Suttorp MJ, Shekelle PG. Spinal manipulative therapy for low-back pain. Cochrane Database of Systematic Reviews. 2004 January; CD000447(1). Available from: http://dx.doi.org/10.1002/14651858.CD000447.pub2
Assendelft WJ, Koes BW, van der Heijden GJ, Bouter LM. The effectiveness of Chiropractic for treatment of low back pain: an update and attempt at statistical pooling. Journal of Manipulative and Physiological Therapeutics. 1996 October;19(8):499–507. Available from: http://view.ncbi.nlm.nih.gov/pubmed/8902660.
Breen A. Manipulation of the neck and stroke: time for more rigorous evidence. Med J Aust 2002; 176:364 -365. Available from: http://www.mja.com.au/public/issues/176_08_150402/bre10066_fm.html.
Breen A. In praise of chiropractic. Journal of Health Services Research & Policy. 2009 July;14(3):188–189. Available from: http://dx.doi.org/10.1258/jhsrp.2009.009025.
Ernst E, Canter PH. A systematic review of systematic reviews of spinal manipulation. J R Soc Med. 2006 April;99(4):192–196. Available from: http://dx.doi.org/10.1258/jrsm.99.4.192.
Evans G, Richards S. Low back pain: an evaluation of therapeutic interventions. University of Bristol, Department of Social Medicine, Health Care Evaluation Unit, 1996:176. (Summary available from: http://www.mrw.interscience.wiley.com/cochrane/cldare/articles/DARE-11996008102/frame.html)
Ferreira ML, Ferreira PH, Latimer J, Herbert R, Maher CG. Efficacy of spinal manipulative therapy for low back pain of less than three months’ duration. Journal of Manipulative and Physiological Therapeutics. 2003;26(9):593–601. Available from: http://dx.doi.org/10.1016/j.jmpt.2003.08.010.
Fisher RA. Design Of Experiments. 1st ed. Edinburgh: Oliver and Boyd; 1935. Reprinted by Oxford University Press.
French SD, Grant W, Green S, Walker BF. Chiropractic interventions for low-back pain (Protocol). Cochrane Database of Systematic Reviews 2005, Issue 3. Art. No.: CD005427. DOI: 10.1002/14651858.CD005427.
Gemmell H, Miller P. Comparative effectiveness of manipulation, mobilisation and the Activator instrument in treatment of non-specific neck pain: a systematic review. Chiropractic & Osteopathy. 2006 April;14(1):7+. Available from: http://dx.doi.org/10.1186/1746-1340-14-7.
Oliphant D. Safety of spinal manipulation in the treatment of lumbar disk herniations: a systematic review and risk assessment. Journal of manipulative and physiological therapeutics. 2004;27(3):197–210. Available from: http://dx.doi.org/10.1016/j.jmpt.2003.12.023.
Pengel HM, Maher CG, Refshauge KM. Systematic review of conservative interventions for subacute low back pain. Clinical Rehabilitation. 2002 December;16(8):811–820. Available from: http://view.ncbi.nlm.nih.gov/pubmed/12501942.
Thanks (I think) to dvnutrix for putting this paper under my nose.
11th July 2009. Comments on Assendelft et al. (2004) added.
11th July 2009. Link added to the post “Alan Breen in praise of chiropractic and democracy” on the Evidence Matters blog.
[BPSDB]
9 Responses to “In praise of chiropractic?”
Sorry, the comment form is closed at this time.
A canna’ change the laws of physics 
dvnutrix said
Always a pleasure (you did not hear a snicker over the aether).
As ever, your discussion of positivism is helpful in setting a context and enlightening.
Nicely expressed:
In the same way, perhaps “democracy and inclusivity” belong in an alt.sociolect in which they mean something other than they appear to mean (NB, I am in line with your interpretation but wouldn’t wish to come across as excluding whatever seems to be excluded by use of the term “inclusivity”).
Even so, Prof. Breen’s partial account of the evidence is disappointing, more so given his own criticism of such partial accounts.
apgaylard said
Thanks. I agree that “democracy and inclusivity” are being invoked to support the use of something that lacks good evidence of efficacy and poses low (but ill-defined) levels of risk. pv (below) makes a reasonable suggestion as to what the implied meaning may be. Other explanations are possible, but the outcome is the same.
I think that the most disappointing things about this article are 1) the lazy pejorative use of “positivist” (quite Milgromesque) and 2) the highly selective use of ‘evidence’.
The last point is the most serious; particularly given that he calls this kind of argument “pseudoscience” (it’s not really, though I concede that many pseudoscience proponents adopt this approach). It’s telling that evidence of efficacy comes from committees, whereas trials only figure in the safety and cost-of-illness discussions.
I might even be persuaded that their is a vested interest at work here 😉
pv said
I agree that “positivism” in this sense is just another word for any science that produces results he doesn’t like. I think also he speaks of “democratic” and “inclusive” to mean that evidence gathering should be much less rigorous and treat statistical anomalies as the equals of near certainties. The results of shoddy trials should be taken to be of equal value as those of double blind RCTs.
But surely the point is that science is neither democratic nor inclusive. It is rightly very undemocratic and exclusive, otherwise it becomes “cargo cult science”, or just “not-science”.
apgaylard said
I could not agree more. Experiment being the final arbiter science could be called a dictdataship 😉 I think that because democracy is (can’t remember who I am quoting here – too early) the least worse form of government, then people make a lazy assumption that its the best way to organize any given activity. This article seems to make that mistake.
I suspect that you are right that “inclusivity” is code for “any shoddy trial I like” (cherry picking).
It could also be that the coments on democracy and inclusivity are getting at an argument for spurious patient choice: we should be free to chose whatever therapy we like within a healthcare system; regardless of risks or benefits.
Thanks for the comment.
gimpy said
Excellent piece Apgaylard, it is interesting that Breen focuses on back pain, the one area where it can be argued there is plausibility in the use of chiropractic (albeit from a musculoskeletal perspective, not arguments based on subluxation) rather than dealing with the claims of implausible applications related to the fuss over Simon Singh’s case. Breen’s article is a nice example of evasion by dealing with criticism of the strongest evidence and avoiding issues arising from the weakest evidence completely. I don’t think it will be long before we see chiropractors citing it in support of their profession as a whole rather than supporting its tacit admission that at best chiropractic is useless other than for back pain.
PS the quote about democracy comes from Winston Churchill.
Hansard, November 11, 1947 apparently
apgaylard said
Gimpy. Thanks – I appreciate the comment. I must admit that I’d never taken the time to look at the evidence around chiropractic and back-pain until this paper fell into my lap: I hadn’t realised that it was so sparse.
I think that you are right to suspect an attempt to validate the whole practise of chiropractic through its apparent influence on back-pain. If only Chiropractic re-branded itself as an exclusively musculoskeletal practice for back-pain, then much of the criticism the profession gets would disappear. Though there would still be the matter of risk vs. benefit – which doesn’t look too good, from what I’ve read at least.
It’s also interesting that Breen’s plea is very much: we have official endorsement so we must be effective. It shows the problems that come with any official recognition of unvalidated practises; including providing a ‘foot in the door’ for the implausible nonsense.
Seeking approval based on the DoH and EC 2005 guidelines may be a bit of a two-edged sword for Chiropractic: these recommend pharmacological pain-control which seems a bit out of step with the ‘drug free’ pitch made by many chiropractors.
Finally, thanks for the Churchill reference.
draust said
Yawm. This is just the Dave Holmes et al. “postmodernism… different ways of knowing” line (again), invoked to excuse a complete lack of reliable evidence FOR implausible therapeutic claims.
Like many scientists far more eminent than I, I prefer the Royal Society’s motto:
“Nullius in verba”
—————————————————
Quite. As Pal MD neatly put it:
apgaylard said
Completely agree with your take on the philosophistry. While I do think that a more evidence-based chiropractic would be better, I must admit to being a bit conflicted. Yes, there is some evidence for it helping with low-back pain, but the evidence isn’t very good, unless I missed something. I think that it’s hard to justify it against any sort of additional risk. I must admit that I was surprised how poor the evidence for chiropractic and low-back pain is.
draust said
Generally the evidence for anything helping with low-back pain is rather poor, AP.
Mrs Dr Aust, who works in a medical specialty where she see a lot of low-back pain, says that the only thing that really helps is getting the person mobile rather than immobile. The cheapest and best thing to get them mobile is paracetamol to deal with pain that makes them not want to get mobile. So the standard approach is something like:
(i) check for (and hopefully exclude) “red-flags” that would indicate something serious is going on;
(ii) give the person sensible advice about mobilising / exercise and the sensible use of pain killers to help with this
plus often:
(iii) refer them to a physio who does specialist back stuff for a few sessions to teach strengthening exercises they can do without “doing” their back again.
Part of the problem for assessing the interventions, and perhaps especially the complementary ones, come from the fact that pain has a large “subjective / perceptual component”, and back pain is no exception. So “placebo therapies” that (in effect) specifically target the belief set of particular patients may do something to help get them up and mobile, though not by acting on anything physical.
Another point is that one of the things many mainstream doctors distrust about CAM practitioners handling back pain is how well they do (i) above. They worry that “green-lighting” CAM (see e.g. the NICE guideline farrago) will mean people being referred – or probably more likely, patients self-referring – to CAM practitioners without a thorough examination and history-taking from a knowledgable conventional doc. Chiropractors in particular try very hard to promote themselves as “fully capable of acting as primary care practitioners for musculoskeletal problems”, but this makes the conventional doctors I know distinctly uneasy.