Non-specific low back pain: Returning to the Dark Ages


While non-specific low back pain (NSLBP) is a common diagnosis, Tim Germon (Derriford Hospital, Plymouth, UK) argues that most low back pain is in fact specific. He speaks about the importance of reaching a diagnosis in order to drive effective treatment, focusing in particular on the often underdiagnosed structural causes of pain, and pain as a manifestation of psychological distress, which, in his opinion, must be treated as a specific diagnosis.

In the sixth century BC Pythagoras had mathematical proof that the sun was at the centre of the solar system and that the Earth was in orbit around it. However, around the beginning of the Common Era, the scientific advancements made by the Greeks were displaced by religious dogma. For 2000 years, people attempted to explain the observed motion of the planets in geocentric terms. This was doomed to fail. It is impossible to explain the observations of the planets by using a model which is fundamentally incorrect. In the 16th century, Copernicus rediscovered what the Greeks had known 2000 years previously.1 It was the dawn of the “Age of Reason” or “Enlightenment”, when people sought rational explanations for the world around us. Huge advances were made in all aspects of science and technology, including medicine. The Hippocratic principle of making a diagnosis on which rational treatment could be based was recognised.

Very few, if any, people disagree that the Hippocratic principle of making a diagnosis should form the foundation of medical practice. This is the fundamental skill of a Doctor of Medicine. Yet, 400 years after the beginning of the Enlightenment, it is permissible, even encouraged, for physicians to use the phrase “non-specific low back pain” (NSLBP) as a diagnosis. No criteria, other than that there is no explanation for the person’s pain, is required. There is no minimum requirement in terms of the expertise of the person making the assessment or of investigations required. It is like returning to the Dark Ages.

Not everyone with low back pain requires a diagnosis. Most pain resolves without intervention. However, if treatment is to be prescribed, then a diagnosis is required. Without a diagnosis there is little rationale for intervention. Only with a meaningful diagnosis can there be discussion of the risks and benefits of potential interventions. Interventions not based on diagnosis risk being dishonest.2 Inadequate attempts at making a diagnosis will mean people for whom effective treatment may be available may not be identified.

Back pain

NSLBP is a label, it is not a diagnosis. Therefore, the rationale for treatment will be questionable and it is likely to be ineffective. This is confirmed by the literature. A meta-analysis of effect sizes for non-surgical treatment of NSLBP confirms effect sizes are what we might expect from a placebo.3 The largest effect size identified in this paper is 0.6 for acupuncture. This compares extremely poorly with total hip replacement, which is one of the most effective pain treatments we have with an effect size of 2.5.4 The same is true of spine surgery. A paper from 2005 concluded that spinal fusion was no more beneficial than intensive rehabilitation.5 It seems bizarre that it was acceptable to operate on people when there was no clear diagnosis.  Perhaps this paper was of its time. However, it remains permissible to perform fusions on people with the label NSLBP as part of a trial according to current UK guidelines.6 Hippocrates would weep.

If we apply basic Hippocratic principles, a person presenting with persistent disabling back pain should see a person trained in making a diagnosis—a doctor. The doctor takes a history, performs an examination and arrives at a differential diagnosis. An MRI scan is overwhelmingly the best test to confirm whether this is the case or not. If a structural cause for pain is confirmed then it is appropriate to discuss the potential risks and benefits of surgical intervention. Surgery, for example to treat nerve root compression, can be extremely effective in treating pain. We have measured the change in pain score following surgery for nerve root compression in 867 unselected consecutive patients. The effect size for treating pain was 2.5, the same as total hip replacement and off the scale compared to conservative treatment.7

But what if there is no apparent diagnosis? What are the possibilities? There are four. 1) The diagnosis has been missed, 2) the person has a condition as yet unrecognised, 3) The person has a central sensitisation syndrome or 4) the person has psychological distress manifesting as pain. Three and four may be different perspectives of the same thing.

As physicians we have to be sure that a structural diagnosis has not been missed. It is my observation that dural compression syndromes, for which extremely effective surgery is available, are commonly underdiagnosed. There are three reasons for this. Firstly, whilst we generally agree that nerve root decompression provides good relief of pain, we cannot agree on the distribution of pain which is a potential manifestation of nerve root compression. We have found that of 123 sequential people undergoing very successful surgery for lumbar nerve root compression, only one fulfilled the diagnostic criteria required by UK national guidelines to diagnose sciatica. The rest would not have been imaged or seen a spine surgeon if these guidelines had been adhered to.8 Secondly, people do not believe that dural compression causes back pain, despite overwhelming evidence to the contrary, with Owens et al contributing the most recent paper to add to the body of evidence.9 Thirdly, nerve root compression is unreliably recognised by radiologists.10

What about the rest? There is little we can do when we really do not know the cause of a person’s pain. But if the patient is suffering from psychological distress or central sensitisation surely we should attempt to make a positive diagnosis? In 1999, three psychiatrists suggested that various functional somatic syndromes across different medical specialties have more in common than they do differences.11 I suggest that back pain for which no structural explanation can be found should be included amongst them.

This is important to consider because making a positive psychiatric or psychological diagnosis is as important as making a structural diagnosis. In 2017, Gittins et al described people labelled with fibromyalgia who were submitted to specialist psychiatric assessment. Eighty per cent were found to have a diagnosis for which there was specific treatment.12 It seems likely that a significant number of people with back pain will be similar.

There are many examples of the medical community being resistant to change despite overwhelming evidence of prevailing management being ineffective. The management of peptic ulceration and cholera, to name but two. Faced with a patient with disabling back pain do we revert to the dark ages, label it non-specific and prescribe witchcraft, or do we continue with enlightened thinking, apply Hippocratic principles, and base rational treatment on a diagnosis? The answer seems obvious and yet the status quo is maintained, perhaps by the vested interests of the NSLBP “industry”? This industry includes both service providers and commissioners and it is built on an extremely unsound foundation.

Tim Germon is a consultant spinal neurosurgeon at Derriford Hospital, Plymouth and is the past President of the British Association of Spine Surgeons (BASS).


  1. Koestler A. The Sleepwalkers. A History of Man’s Changing Vision of the Universe. Hutchinson, London 1959.
  2. Germon T, Clifford D, Lee W & Hobart J. The Lancet low back pain series. Lancet (in press).
  3. Keller, A., et al. “Effect sizes of non-surgical treatments of non-specific low-back pain.”European Spine Journal (2007);16.11: 1776-1788.
  4. Ostendorf, M., et al. “Patient-reported outcome in total hip replacement: a comparison of five instruments of health status.” The Journal of bone and joint surgery. British volume (2004);86.6:801-808.
  5. Fairbank, Jeremy, 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);7502: 1233.
  7. Phang Y, Hobart J and Germon T. Spinal nerve root decompression is as effective as total hip replacement (THR) for treating pain. Brit J Neurosurg (2018);32(3):313.
  8. Germon, Tim, William Singleton, and Jeremy Hobart. “Is NICE guidance for identifying lumbar nerve root compression misguided?” European Spine Journal (2014); 23.1: 20-24.
  9. Owens II, R. Kirk, et al. “Back pain improves significantly following discectomy for lumbar disc herniation.” The Spine Journal(2018).
  10. Hussan, F., et al. “Discrepancy in surgical and radiological reporting of lumbo-sacral MRI.”Orthopaedic Proceedings. Vol. 91. No. SUPP III. Orthopaedic Proceedings, 2009.
  11. Wessely, Simon, Chaichana Nimnuan, and Michael Sharpe. “Functional somatic syndromes: one or many?” The Lancet (1999);354.9182:936-939.
  12. Gittins, Rosalind, et al. “The accuracy of a fibromyalgia diagnosis in general practice.” Pain Medicine(2017);19.3: 491-498.


Please enter your comment!
Please enter your name here