Back Pain: What Guidelines Actually Recommend

The gap between evidence and practice here is enormous — and most of it involves things people are sold rather than things they do.

What this is

Low back pain is one of the leading causes of disability worldwide, and one of the most over-treated conditions in medicine. The Lancet's 2018 series on it found a consistent international pattern: limited use of the recommended first-line treatments, alongside inappropriately high use of imaging, rest, opioids, spinal injections and surgery.

The recommended approach is comparatively undramatic. UK guidance (NICE NG59) advises risk stratification, education supporting self-management, staying active, exercise as a core treatment, and prudent use of medication, imaging and surgery.

What the evidence says

  • Exercise is a core recommended treatment

    Gold· human data

    Endorsed in NICE guidance and international guidelines. The type matters far less than doing it — no single exercise modality has proven decisively superior.

  • Staying active beats rest

    Gold· human data

    Prolonged rest worsens outcomes. Continuing normal activity as far as pain allows is the standard advice.

  • Routine imaging does not improve outcomes and may worsen them

    Gold· human data

    Guidelines advise against imaging in non-specific back pain without red flags. Disc changes are common in people without any pain, so a scan often finds something incidental that then gets treated.

  • Psychological factors influence outcome

    Gold· human data

    Fear of movement, distress and expectations predict persistence, which is why guidelines use a biopsychosocial framework rather than a purely mechanical one.

  • Posture correction and "alignment" fix back pain

    Unproven

    Posture correlates poorly with pain, and posture-correction products have little to support them. Strength, activity and confidence in moving have better evidence.

  • Opioids, injections and surgery for ordinary back pain

    Not supported

    The Lancet series named these as overused low-value care in non-specific back pain. There are specific indications; ordinary mechanical back pain isn't one.

What this means in practice

  • Keep moving within tolerance, and expect most episodes to settle over weeks.
  • Do exercise you'll continue. Progressive strength work has good support; so does walking, swimming, Pilates and yoga.
  • Skip the scan unless there are red flags or a clinician has a specific reason.
  • Ignore posture-corrector products; build strength and tolerance instead.
  • If pain persists beyond six weeks or is escalating, see a clinician about stratified management rather than repeating what hasn't worked.

When to see a doctor

  • Loss of bladder or bowel control, or numbness in the saddle area — emergency assessment.
  • Progressive weakness in one or both legs.
  • Back pain with fever, night sweats or unexplained weight loss.
  • New severe back pain over 50, with a history of cancer, or after significant trauma.

The honest summary

Stay active, exercise, avoid the scan, avoid the posture gadget, and get red flags checked immediately. The evidence-based path is unexciting, which is precisely why so much of the market points elsewhere.

Work it out

See also

Frequently asked questions

Do I need an MRI for back pain?
Usually not. Guidelines advise against routine imaging in non-specific back pain without red flags, partly because incidental findings are so common in people with no pain at all — and finding one can lead to treatment you didn't need.
What's the best exercise for back pain?
The one you'll keep doing. No modality has proven decisively superior; strength training, walking, Pilates and yoga all have supportive evidence. Consistency and gradual progression do the work.
Is bad posture causing my back pain?
Posture correlates poorly with pain, and posture-correction products have little evidence. Strength, activity levels and confidence in moving matter more.

Sources

Written and reviewed by Mathew Beale, MSc Biotechnology, University of Reading.

Last reviewed: