By Edzard Ernst
Chiropractic has not often been out of the news in recent months – bad news, that it! It started with the weird decision of the British Chiropractic Association to sue my friend and co-author (TRICK OR TREATMENT) Simon Singh. They not only lost the case but also money and reputation. What followed was dramatic: over 700 chiropractors were investigated by the General Chiropractic Council for making bogus claims.
As a result of all of this, most chiropractors do no longer advertise their services for non-spinal conditions, at least not openly. The more sensible amongst their ranks realise that neck and back-pain is their domain. The trouble is, however, that for neck pain in particular, the evidence is not that convincing either. Therefore a new trial was greeted with loud applause.
Within hours of the publication of this new US study1, the world of chiropractic celebrated it as a vindication of chiropractic spinal manipulation. Its aim was to “to determine the relative efficacy of spinal manipulation therapy (SMT), medication, and home exercise with advice (HEA) for acute and subacute neck pain in both the short and long term”. Because neck pain is such a common problem which is often difficult to treat, this study did indeed seem important.
The researchers recruited 272 patients suffering from nonspecific neck pain since 2 to 12 weeks, subsequently they treated them for 12 weeks with either SMT, medication, or HEA. The primary outcome was pain which was measured at 2, 4, 8, 12, 26, and 52 weeks. The results suggested that SMT had a significant advantage over medication after 8, 12, 26, and 52 weeks, and HEA was superior to medication at 26 weeks. No clinically important differences were found between SMT and HEA at any time. The authors concluded that, “for participants with acute and subacute neck pain, SMT was more effective than medication in both the short and long term. However, a few instructional sessions of HEA resulted in similar outcomes at most time points”.
At first glance, this seems to be a rigorous piece of research. At closer scrutiny, however, the flaws of this study become fairly obvious. Here are the ones that strike me as particularly important. The medication group was treated in a different setting than the other two groups. The drugs administered were not clearly defined, and no information was provided about the dosage or the length of the drug therapy. Six patients of the medication group received no treatment at all. Thus any comparisons between patients receiving medication and the other two groups are problematic, to say the least.
Similar problems exist regarding the comparison between the SMT and the exercise group. There was no adequate control for non-specific effects. It is obvious that the therapeutic encounter and “tender loving care” (TLC) can be beneficial for patients who suffer from neck pain. The SMT group enjoyed an average of 15 hands-on sessions of TLC while the exercise group had just 2 sessions of instructions. For this reason alone, the comparison between these two groups tells us next to nothing about the specific effects of chiropractic SMT.
I therefore fear that this study merely shows that TLC and non-specific effects can strongly influence symptoms such as pain. If we consider the high costs of regular SMT versus the negligible expense of HEA, the latter would probably turn out to be preferable. If we finally factor in the potential for harm, the balance decidedly tilts towards HEA. Exercise is virtually risk-free, whereas SMT results in transient adverse effects in about 50% of all patients2; in addition, it is associated with several hundred severe complications including deaths3.
My conclusion is simple: Chiropractic, or more accurately, the uncritical promotion of this approach, can therefore be a pain in the neck.
(1) Bronfort G, Evans R, Anderson AV, Svendsen KH, Bracha Y, Grimm RH. Spinal manipulation, medication or home exercise with advice for acute and subacute neck pain. A randomised trial. Ann Intern Med 2012; 156:1-10.
(2) Stevinson C, Ernst E. Risks associated with spinal manipulation. Am J Med 2002; 112:566-570.
(3) Ernst E. Deaths after chiropractic: a review of published cases. Int J Clin Pract 2010; 64(8):1162-1165.