MD, PhD, FMedSci, FRSB, FRCP, FRCPEd

Chiropractors, like other alternative practitioners, use their own unique diagnostic tools for identifying the health problems of their patients. One such test is the Kemp’s test, a manual test used by most chiropractors to diagnose problems with lumbar facet joints. The chiropractor rotates the torso of the patient, while her pelvis is fixed; if manual counter-rotative resistance on one side of the pelvis by the chiropractor causes lumbar pain for the patient, it is interpreted as a sign of lumbar facet joint dysfunction which, in turn would be treated with spinal manipulation.

All diagnostic tests have to fulfil certain criteria in order to be useful. It is therefore interesting to ask whether the Kemp’s test meets these criteria. This is precisely the question addressed in a recent paper. Its objective was to evaluate the existing literature regarding the accuracy of the Kemp’s test in the diagnosis of facet joint pain compared to a reference standard.

All diagnostic accuracy studies comparing the Kemp’s test with an acceptable reference standard were located and included in the review. Subsequently, all studies were scored for quality and internal validity.

Five articles met the inclusion criteria. Only two studies had a low risk of bias, and three had a low concern regarding applicability. Pooling of data from studies using similar methods revealed that the test’s negative predictive value was the only diagnostic accuracy measure above 50% (56.8%, 59.9%).

The authors concluded that currently, the literature supporting the use of the Kemp’s test is limited and indicates that it has poor diagnostic accuracy. It is debatable whether clinicians should continue to use this test to diagnose facet joint pain.

The problem with chiropractic diagnostic methods is not confined to the Kemp’s test, but extends to most tests employed by chiropractors. Why should this matter?

If diagnostic methods are not reliable, they produce either false-positive or false-negative findings. When a false-negative diagnosis is made, the chiropractor might not treat a condition that needs attention. Much more common in chiropractic routine, I guess, are false-positive diagnoses. This means chiropractors frequently treat conditions which the patient does not have. This, in turn, is not just a waste of money and time but also, if the ensuing treatment is associated with risks, an unnecessary exposure of patients to getting harmed.

The authors of this review, chiropractors from Canada, should be praised for tackling this subject. However, their conclusion that “it is debatable whether clinicians should continue to use this test to diagnose facet joint pain” is in itself highly debatable: the use of nonsensical diagnostic tools can only result in nonsense and should therefore be disallowed.

11 Responses to How reliable are chiropractic diagnostic tests?

  • Can I point out that Kemp’s Test is used by all manual therapist (osteopath, physiotherapist and chiropractor) and orthopaedic surgeons – so not really a chiropractic diagnostic test but an orthopaedic diagnostic test

    • This is quite correct, this test sometimes called a quadrant test is used extensively by physiotherapists. Recently, i have seen this used diagnostically in pain clinics prior to facet joint injection. This is despite facet joint injections not being supported by evidence and not recommended in NICE guidelines.

      The most important part of this post is the praise given to chiropractors (maybe reluctantly) for highlighting the weakness of this tes. As others have stated reliance on a single test is inappropriate regardless of your profession

      • I agree with Andy, an important part of Prof. Ernst article is the praise for tackling the subject of sensitivity and specificity of orthopaedic tests. However, the overall article could and maybe should have been written from an objective view point.

  • Can I also point out that this article is biased too

    • In what way is EE’s article biased?

      • Re: Irene

        The article is biased because Prof. Ernst took a recent paper on a orthopaedic test (see above) used by all professions and went on to imply that it is a unique test used by most chiropractors (it is not). He then uses the conclusion from that recent paper to criticize chiropractors for treating conditions which the patient does not have.

        Prof Ernst then must not realise that most orthopaedic test have issues with specificity and sensitivity and is therefore a issue for all professions that use them (i.e. physios, osteos, chiros and orthopaedic surgeons). For him to write the above article criticizing just one profession shows a lack of knowledge on the topic and also a bias against chiropractors.

        However, if he did an article on the neurocalometer, I would agree with him that this is a dubious diagnostic device specifically used by chiropractors. But he didn’t and therefore this article should be criticized for what it is – a poor / misguided attempt to malign a profession by using a recent research paper on a orthopaedic test which is used by all professions.

  • Just my personal opinion, but it’s such a pity that chiropractors and osteopaths can’t do a good springclean job on their diagnostic methods and claims. I know I’m not alone in having had successful treatment from both: walked into osteopath’s room with really painful lower back, 25 minutes later walked out pain free, and it didn’t return. Went to a chiropractor 4 times after 18 months of worsening sciatica, he almost completely cured it. It seems they learn a great deal, and almost certainly help a lot of people, if only they’d just drop the outlandish claims and subject their methods to evidence-based science.

  • If anyone uses just one test to make a diagnosis they should be hung, drawn and quartered!
    There is definitely bias in the tags! This post should be tagged chiro, physio, medical and orthopedic minimum to cover all who use the test. If this paper was authored by a physio or orthopod would it have even made it onto this blog.

    I find this surprising for someone as thorough as you Edzard, are you intentionally stiring the pot to see who bites??

  • The above article by Prof. Ernst is so wrong, in so many different ways.

  • This strikes a chord with me. Many years ago, I had a course of treatment from a chiropractor for a shoulder problem. Two tests she did each week to decide how to treat me during that session struck me as woefully inaccurate.

    After I was lying on her bench, she’d ask me to try to raise both legs – how easy I found it partly determined which bit of my body she thought needed treatment that session. I soon realised that what was determining my ability to raise my legs was simply the angle of my pelvis as I lay down – if it was tucked under, I could raise my legs easily, but if my back was slightly arched, I couldn’t. So the position of my pelvis was pure chance, dependent on how I’d lain down on the bench, and how the flesh and muscles of my buttocks had settled on it. It was useless as a diagnostic test.

    My shoulder was frozen, and one procedure she’d do on it was helpful in freeing it up a little. But first she would ask me to make a circle with my thumb and index finger, and she’d put her fingers into the circle and try to pull my fingers apart while telling me to try and stop her. If she could prise my fingers apart, she’d treat the shoulder – if she couldn’t, she wouldn’t treat it. The stupid thing was that she knew I had RSI in my wrists and forearms, so whether I was able to stop her pulling my fingers apart depended on how bad my RSI was that day, ie if I’d been using the computer a lot.

    Eventually, I got wise to this test and always let her pull my fingers apart, because the procedure she’d then do on my shoulder was the most useful thing in the entire session, the thing I’d gone to her for in the first place. She thought I had other problems and was treating those as well as the shoulder. In the case of the neck, I ended up with worsening headaches over the months I went to her because of the way she seemed to be adjusting the carriage of my head. The headaches were the reason I stopped going to her, even though she had helped loosen the shoulder.

    When I first went to her, I took the MRI scans the consultant had had done on the shoulder. She looked at them for a few moments and then said, ” Well, of course you can’t tell much from them.” Actually, you could tell a great deal from them, since they showed the soft tissue so clearly – the consultant had used them to determine which tendon was inflamed. But she was too ignorant to know that, and thought her diagnostic tests, including the two I’ve described, were more accurate.

  • Excellent post. Very scientific, very artist, veryvery good. Thank you Prof Ernst.

Leave a Reply

Your email address will not be published. Required fields are marked *

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted.


Click here for a comprehensive list of recent comments.

Categories