Alexander technique or exercise for treatment of back pain

20th February 2009

The treatment of chronic back pain remains challenging. Up to 70% of the population will experience back pain at some stage but only about 7% have symptoms for more than two weeks and only 1% require long term treatment (Devereaux et al. Med Clin North Am 2003; 87: 643-662). The biggest single problem is that in up to 70% of patients a firm diagnostic “label” is never achieved. Many patients and treating practitioners find this difficult to accept. It is difficult to conceptualise and implement a rational and systematic treatment when the underlying cause remains unknown!

The devotees of the Alexander technique will have been encouraged by the recent publication by Little et al in the BMJ (BMJ 2008; 337: a884). This is the first time that the technique has been shown to be of any benefit in back pain treatment. This study compared 6 or 24 Alexander sessions with massage and “normal” care in 579 patients in 64 general practices in England. The results showed that the Alexander patients had significantly less pain and better function at 1 year compared to the other groups: 6 sessions plus exercise were almost as effective as 24.

Inevitably, as with any paper, there are a number of problems. The study patients were randomly selected from those who had attended their GP with chronic back pain in the previous 5 years but were NOT seeking treatment at the time of inclusion in the study. This group is probably NOT comparable to those most of us are really concerned with: those attending their GP seeking help!

One might also reasonably argue that the Alexander patients had relative success compared to two other treatments which we already know are poor! A better and more valid comparison would have been some other form of exercise therapy where some efficacy has already been demonstrated (Cochrane database 2005).

My take on the issue is that further and more convincing evidence is required before the technique is accepted as a proven treatment for back pain. It is incumbent on the proponents of the treatment to provide this. The evidence provided in the above paper is, at best, weak and stands in isolation. This is a technique after all that was originally designed to allow Alexander to improve his oratorical skills and has been advocated for treatments of other illnesses such as Parkinson’s disease (Stallibrass, Clin Rehab 2002;17(7):695-708) . The jury is still out and many remain sceptical.

What about exercise therapy? NICE is currently in the process of consulting on its draft guidelines on the acute management of chronic (longer than 6 weeks) non specific, low back pain (available on their website). Their review captured 1195 papers relating to exercise programmes from which they examined 59. Only seven were finally included.

Hayden et al (Cochrane Database of Systematic Reviews, Issue 3 2006) performed a high quality, systematic review looking at exercise in subacute and chronic low back pain compared to no treatment or other conservative treatment. There was insufficient evidence in relation to subacute back pain to draw conclusions either for or against. In the chronic group, exercise was found to be slightly effective at decreasing pain and increasing function.

The UK BEAM trial (BMJ 2004; 329 (7479): 1377-1384) examined exercise and/or manipulation in addition to standard care in 1334 patients. The exercise group had statistically significant improvements in disability scores, SF36 physical scores and fear avoidance physical scores at 3 months only. Pain scores were improved at 3 and 12 months. Both of these papers are of high quality and low risk of bias and should be of interest particularly to physios with an interest in back pain.

A Finnish study (Kuukkanen et al. Physiotherapy Res Int 2007; 12 (4): 213-224) concluded that supervised controlled home exercises reduced low back pain lasting over 5 years. A Swiss study (Maul et al. Eur Spine J 2005; 14 (6): 599-611) showed that supervised physical training improved functional capacity in both muscular endurance and isokinetic strength. Self rated pain and disability were significantly decreased at one year. Whilst of interest, both papers have a high risk of bias. Two other included studies examined hydrotherapy and yoga treatments.

Following examination of the relevant health economic evidence (Hayden et al. Ann Inter Med 2005; 142 (9): 776-785 and Mannion et al. Rheumatology 2001; 40 (7): 772-778)) the NICE draft recommendations in relation to exercise are as follows:

  • There is evidence for clinical effectiveness of structured exercise programmes in terms of improved function and reduced disability. The size of these effects is generally small and there was no evidence of improved psychological distress.
  •  Exercise alone appears cost effective compared to best care in general practice. The addition of manipulation improves results yet further but is not always available.
  • There is no evidence that one to one based exercise is better than group exercise, the latter being preferred as more cost effective.

(The guidelines deal with all other aspects of care and are worth reading!)

It remains the case therefore that there is no easy, quick fix. The grail remains an understanding of the causes of non specific low back pain hopefully followed by sensible and scientifically valid treatment.


Tony Reece FRCS Tr&Orth
Consultant Spinal Surgeon
November 2008


« back to back & neck