This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

A faulty production line

NICE dramatically reduces drug options for low back pain

GPs should not prescribe TCAs, SSRIs or strong opioids for lower back pain, and only offer paracetamol for second-line use, under planned revisions to NICE guidelines unveiled today.

The draft guidelines – which now also cover sciatica – say GPs should offer NSAIDs as first-line for pain relief, and should offer paracetamol only alongside a weak opioid. 

The guidelines also say GPs should avoid acupuncture altogether – which they say is no better than sham treatment – and call for exercise, such as stretching, strengthening, aerobics or yoga, to be the first step to help patients manage their condition.

The guidance also states massage and manipulation by a therapist should only be offered alongside exercise.

The proposed recommendations downgrade the use of paracetamol, which should no longer be offered first-line for pain relief, or used on its own. Instead GPs should suggest patients try an NSAID such as ibuprofen or aspirin first.

GPs should only consider using weak opioids such as codeine – which may be given with or without paracetamol – if patients cannot tolerate an NSAID, or find they do not work. Stronger opioids are completely ruled out.

The guidance calls for GPs to use a risk assessment and stratification tool such as Keele University’s STarT Back tool to help make a decision with the patient on the best course of management, depending on the severity of the condition.

And it says they should consider offering a combined physical and psychological programme – preferably in a group – for people with ‘significant psychosocial obstacles to recovery’.

GPs should avoid imaging if possible, however, and the guidelines rule out use of electrotherapies such as TENS (transcutaneous electrical nerve stimulation), and spinal injections.

But GPs can consider referral for radiofrequency denervation in patients with moderate to severe pain that has not responded to other treatment, and epidural steroid injections for people with acute sciatica.

Of potential surgical interventions, only spinal decompression is recommended – in people with sciatica that has not responded to non-surgical approaches.

Previously GPs could offer a course of 10 sessions of acupuncture, or manual therapy, as alternative options to an exercise programme. However, NICE guidelines advisors say the most up-to-date evidence shows that acupuncture is no better than sham treatment and that there is no evidence to support physiotherapy on its own.

Dr Ian Bernstein, a GP expert in musculoskeletal therapy who advised on the updated guidelines, says: ‘The diagnosis of back pain includes a variety of patterns of symptoms.

‘This means that one approach to treatment doesn’t fit all. Therefore the draft guidance promotes combinations of treatments such as exercise with manual therapy or combining physical and psychological treatments, and the choices made should take into account people’s preferences as well as clinical considerations.’

NICE Draft CG88 Update - Non-specific low-back pain and sciatica: management

NICE draft guidelines on management of low back pain and sciatica - key points for GPs

Assessment and diagnosis

GPs should consider carrying out risk stratification (for example, the STarT Back risk assessment tool) when a patient first presents with non-specific low back pain with or without sciatica, ‘to inform shared decision-making about stratified management’.

GPs should not routinely offer scans in a non-specialist setting, and explain to patients that referral to a specialist does not necessarily mean they need to have a scan


  • Do not offer acupuncture, but offer patients a suitable group exercise programme, including biomechanical, aerobic, mind-body exercises or a combination of these.
  • Consider manipulation and massage – but only alongside exercise as part of multimodal treatment
  • Prescibe an NSAID for managing pain where possible – but at the lowest dose possible for the shortest possible amount of time
  • Only offer paracetamol second line, and in combination with a ‘weak’ opioid like codeine
  • Other opioids should not be prescribed for low back pain, nor antidepressants such as SSRIs, SNRIs or TCAs, or anticonvulsants

Readers' comments (66)

  • It is very heartening to see the critical assessment of evidence in this guidance (in contrast to the 2009 guidance). The fact is that none of the treatments works very well and that has led to clutching at straws. This guidance provides a welcome debunking of many of the myths.
    Not least, it may herald the end of acupuncture provision on the NHS.

    I blogged the new guidance this morning, at

    Unsuitable or offensive? Report this comment

  • I am surprised at the many natural non-drug treatments for back pain that NICE has ignored, where there is good evidence.

    These include Vitamin D & Magnesium to improve muscle strength and Omega-3 to reduce the Inflammation. For all 3 products, therapeutic doses are required and not trivial OTC amounts


    Unsuitable or offensive? Report this comment

  • This comment has been moderated

  • The last post was sarcasm I hope ?

    Unsuitable or offensive? Report this comment

  • I fear that some of the comments here illustrate very well why NICE.
    The results of more than 3000 trials of acupuncture, some of them well-designed, have consistently shown that it is indistinguishable from various sorts of sham, yet some people still advocate it. The power of myths is indeed strong (especially when there are no good alternatives). The evidence presented in Anesthesia and Analgesia (2013) shows that it simply does not work (-see ).

    In fact not only does it not work, it doesn't even have a very big placebo effect in most cases. Most of the apparent effects are not real placebo effects. Rather they can be explained regression to the mean (and so of no benefit to the patient). See

    Unsuitable or offensive? Report this comment

  • Unless there are red flags the best treatment for back pain is nothing. Unless of course the pain is severe enough to warrant analgesia, in which case NSAIDs are sensible, because there is a lot of inflammation locally. But as 95% of back pain sufferers get better within 6 weeks anyway (in the case of my recent disc prolapse, significantly better in just 10 days) any so-called treatment is almost certain to "work" - because doing nothing works. The difficulty is in persuading patients that nothing is a positive treatment. So far as investigations are concerned they serve no purpose whatever with mechanical back pain. Find a narrow disc on X-ray? You have no idea how long it's been there. Find a disc bulge on MRI? You aren't going to do anything about it, so as it won't alter your treatment why bother with the scan?

    Of course I have been saying all of this for over 20 years, so NICE is actually a bit behind the times...

    Unsuitable or offensive? Report this comment

  • Another example of the madness of the patient not paying the doctor
    In countries where the patient pays his doctor ( that's everywhere but the UK) doctors who do not help their patients do not eat
    God give me strength to endure the madness of the NHS

    Unsuitable or offensive? Report this comment

  • Vinci Ho

    Interesting debate so far:
    (1) I totally agree that none of the 'options' of treatment really change the course and prognosis of this 'back pain syndrome', even in many cases of clear-cut disc prolapse .(considering I am still a member of BMAS)
    (2) Once the red flags are excluded , those longstanding non specific back pain represents the classical interesection of the three big domains of physical , mental and social problems. We simply do not have a reliable solution in general practice .
    (3) Ten minutes appointment to deal with these patients is diabolical , which always lead to my long advocating in-house physiotherapist ideally in each practice.
    (4) Not surprisingly , some quick fix ,purely symptomatic relief is always demanded by these desperate patients with 'agony'. There is certainly an explosion of usage of opiates as well neuropathic pain drugs. I did not have Butran patch ,lidocaine patch, gabapentin , pregabalin , tramadol etc to prescribe when I started in general practice in 1995. The philosophical question is 'have we significantly improved the care of these patients and why?'
    (5) Perhaps there is a need in changing the 'culture' of understanding and perceiving 'chronic back pain' . What about equating that to chronic headache?

    Unsuitable or offensive? Report this comment

  • Vinci Ho

    Any evidence of medication-induced back pain (instead of headache)??

    Unsuitable or offensive? Report this comment

  • another peace of guidance from NICE to ignore

    Unsuitable or offensive? Report this comment

  • Strong opiates are good in acute slipped disc. if patient is screaming and cannot stand, would you use this Nice guidance? What a really good example of thinking by committee. i deal with a lot of addicts and none of them sourced their syrong opiates for lower back pain - the mental and emotional pain (untreated) came first. Also, whatever happened to involving the patient in the decision - if their belief in acupuncture or physio works for them, isnt that important?!

    Unsuitable or offensive? Report this comment

View results 10 results per page20 results per page50 results per page

Have your say